PMCC PMCC

Search tips
Search criteria

Advanced
Results 1-25 (41)
 

Clipboard (0)
None
Journals
Year of Publication
1.  Using machine learning algorithms to guide rehabilitation planning for home care clients 
Background
Targeting older clients for rehabilitation is a clinical challenge and a research priority. We investigate the potential of machine learning algorithms – Support Vector Machine (SVM) and K-Nearest Neighbors (KNN) – to guide rehabilitation planning for home care clients.
Methods
This study is a secondary analysis of data on 24,724 longer-term clients from eight home care programs in Ontario. Data were collected with the RAI-HC assessment system, in which the Activities of Daily Living Clinical Assessment Protocol (ADLCAP) is used to identify clients with rehabilitation potential. For study purposes, a client is defined as having rehabilitation potential if there was: i) improvement in ADL functioning, or ii) discharge home. SVM and KNN results are compared with those obtained using the ADLCAP. For comparison, the machine learning algorithms use the same functional and health status indicators as the ADLCAP.
Results
The KNN and SVM algorithms achieved similar substantially improved performance over the ADLCAP, although false positive and false negative rates were still fairly high (FP > .18, FN > .34 versus FP > .29, FN. > .58 for ADLCAP). Results are used to suggest potential revisions to the ADLCAP.
Conclusion
Machine learning algorithms achieved superior predictions than the current protocol. Machine learning results are less readily interpretable, but can also be used to guide development of improved clinical protocols.
doi:10.1186/1472-6947-7-41
PMCID: PMC2235834  PMID: 18096079
2.  Models predicting the growth response to growth hormone treatment in short children independent of GH status, birth size and gestational age 
Background
Mathematical models can be used to predict individual growth responses to growth hormone (GH) therapy. The aim of this study was to construct and validate high-precision models to predict the growth response to GH treatment of short children, independent of their GH status, birth size and gestational age. As the GH doses are included, these models can be used to individualize treatment.
Methods
Growth data from 415 short prepubertal children were used to construct models for predicting the growth response during the first years of GH therapy. The performance of the models was validated with data from a separate cohort of 112 children using the same inclusion criteria.
Results
Using only auxological data, the model had a standard error of the residuals (SDres), of 0.23 SDS. The model was improved when endocrine data (GHmax profile, IGF-I and leptin) collected before starting GH treatment were included. Inclusion of these data resulted in a decrease of the SDres to 0.15 SDS (corresponding to 1.1 cm in a 3-year-old child and 1.6 cm in a 7-year old). Validation of these models with a separate cohort, showed similar SDres for both types of models. Preterm children were not included in the Model group, but predictions for this group were within the expected range.
Conclusion
These prediction models can with high accuracy be used to identify short children who will benefit from GH treatment. They are clinically useful as they are constructed using data from short children with a broad range of GH secretory status, birth size and gestational age.
doi:10.1186/1472-6947-7-40
PMCID: PMC2246105  PMID: 18076760
3.  Effect of electronic patient record use on mortality in End Stage Renal Disease, a model chronic disease: retrospective analysis of 9 years of prospectively collected data 
Background
In chronic disease, health information technology promises but has yet to demonstrate improved outcomes and decreased costs. The main aim of the study was to determine the effects on mortality and cost of an electronic patient record used in daily patient care in a model chronic disease, End Stage Renal Disease, treated by chronic maintenance hemodialysis. Dialysis treatment is highly regulated, and near uniform in treatment modalities and drugs used.
Methods
The particular electronic patient record, patient-centered and extensively coded, was used first in patient care in 3 dialysis units in New York, NY in 1998, 1999, and 2000. All data were stored "live"; none were archived. By December 31, 2006, the patients had been treated by maintenance hemodialysis for a total of 3924 years. A retrospective analysis was made using query tools embedded in the software. The United States Renal Data System dialysis population served as controls. In all there were 1790 patients, with many underlying primary diseases and multiple comorbid conditions affecting many organ systems. Year by year mortality, hospital admissions, and staffing were analyzed, and the data were compared with national data compiled by the United States Renal Data System.
Results
Analyzed by calendar year after electronic patient record implementation, mortality decreased strikingly. In years 3–9 mortality was lower than in years 1–2 by 23%, 48%, and 34% in the 3 units, and was 37%, 37%, and 35% less than that reported by the United States Renal Data System. Clinical staffing was 25% fewer per 100 patients than the national average, thereby lowering costs.
Conclusion
To our knowledge, this is the first demonstration that an electronic patient record, albeit of particular design, can have a favorable effect on outcomes and cost in chronic disease. That the population studied has many underlying diseases affecting all organ systems suggests that the electronic patient record design may enable application to many fields of medical practice.
doi:10.1186/1472-6947-7-38
PMCID: PMC2238736  PMID: 18045495
4.  The effect of attitude to risk on decisions made by nurses using computerised decision support software in telephone clinical assessment: an observational study 
Background
There is variation in the decisions made by telephone assessment nurses using computerised decision support software (CDSS). Variation in nurses' attitudes to risk has been identified as a possible explanatory factor. This study was undertaken to explore the effect of nurses' attitudes to risk on the decisions they make when using CDSS. The setting was NHS 24 which is a nationwide telephone assessment service in Scotland in which nurses assess health problems, mainly on behalf of out-of-hours general practice, and triage calls to self care, a service at a later date, or immediate contact with a service.
Methods
All NHS 24 nurses were asked to complete a questionnaire about their background and attitudes to risk. Routine data on the decisions made by these nurses was obtained for a six month period in 2005. Multilevel modelling was used to measure the effect of nurses' risk attitudes on the proportion of calls they sent to self care rather than to services.
Results
The response rate to the questionnaire was 57% (265/464). 231,112 calls were matched to 211 of these nurses. 16% (36,342/231,112) of calls were sent to self care, varying three fold between the top and bottom deciles of nurses. Fifteen risk attitude variables were tested, including items on attitudes to risk in clinical decision-making. Attitudes to risk varied greatly between nurses, for example 27% (71/262) of nurses strongly agreed that an NHS 24 nurse "must not take any risks with physical illness" while 17% (45/262) disagreed. After case-mix adjustment, there was some evidence that nurses' attitudes to risk affected decisions but this was inconsistent and unconvincing.
Conclusion
Much of the variation in decision-making by nurses using CDSS remained unexplained. There was no convincing evidence that nurses' attitudes to risk affected the decisions made. This may have been due to the limitations of the instrument used to measure risk attitude.
doi:10.1186/1472-6947-7-39
PMCID: PMC2238735  PMID: 18047658
5.  Creating a medical dictionary using word alignment: The influence of sources and resources 
Background
Automatic word alignment of parallel texts with the same content in different languages is among other things used to generate dictionaries for new translations. The quality of the generated word alignment depends on the quality of the input resources. In this paper we report on automatic word alignment of the English and Swedish versions of the medical terminology systems ICD-10, ICF, NCSP, KSH97-P and parts of MeSH and how the terminology systems and type of resources influence the quality.
Methods
We automatically word aligned the terminology systems using static resources, like dictionaries, statistical resources, like statistically derived dictionaries, and training resources, which were generated from manual word alignment. We varied which part of the terminology systems that we used to generate the resources, which parts that we word aligned and which types of resources we used in the alignment process to explore the influence the different terminology systems and resources have on the recall and precision. After the analysis, we used the best configuration of the automatic word alignment for generation of candidate term pairs. We then manually verified the candidate term pairs and included the correct pairs in an English-Swedish dictionary.
Results
The results indicate that more resources and resource types give better results but the size of the parts used to generate the resources only partly affects the quality. The most generally useful resources were generated from ICD-10 and resources generated from MeSH were not as general as other resources. Systematic inter-language differences in the structure of the terminology system rubrics make the rubrics harder to align. Manually created training resources give nearly as good results as a union of static resources, statistical resources and training resources and noticeably better results than a union of static resources and statistical resources. The verified English-Swedish dictionary contains 24,000 term pairs in base forms.
Conclusion
More resources give better results in the automatic word alignment, but some resources only give small improvements. The most important type of resource is training and the most general resources were generated from ICD-10.
doi:10.1186/1472-6947-7-37
PMCID: PMC2267171  PMID: 18036221
6.  A comparative analysis of predictive models of morbidity in intensive care unit after cardiac surgery – Part II: an illustrative example 
Background
Popular predictive models for estimating morbidity probability after heart surgery are compared critically in a unitary framework. The study is divided into two parts. In the first part modelling techniques and intrinsic strengths and weaknesses of different approaches were discussed from a theoretical point of view. In this second part the performances of the same models are evaluated in an illustrative example.
Methods
Eight models were developed: Bayes linear and quadratic models, k-nearest neighbour model, logistic regression model, Higgins and direct scoring systems and two feed-forward artificial neural networks with one and two layers. Cardiovascular, respiratory, neurological, renal, infectious and hemorrhagic complications were defined as morbidity. Training and testing sets each of 545 cases were used. The optimal set of predictors was chosen among a collection of 78 preoperative, intraoperative and postoperative variables by a stepwise procedure. Discrimination and calibration were evaluated by the area under the receiver operating characteristic curve and Hosmer-Lemeshow goodness-of-fit test, respectively.
Results
Scoring systems and the logistic regression model required the largest set of predictors, while Bayesian and k-nearest neighbour models were much more parsimonious. In testing data, all models showed acceptable discrimination capacities, however the Bayes quadratic model, using only three predictors, provided the best performance. All models showed satisfactory generalization ability: again the Bayes quadratic model exhibited the best generalization, while artificial neural networks and scoring systems gave the worst results. Finally, poor calibration was obtained when using scoring systems, k-nearest neighbour model and artificial neural networks, while Bayes (after recalibration) and logistic regression models gave adequate results.
Conclusion
Although all the predictive models showed acceptable discrimination performance in the example considered, the Bayes and logistic regression models seemed better than the others, because they also had good generalization and calibration. The Bayes quadratic model seemed to be a convincing alternative to the much more usual Bayes linear and logistic regression models. It showed its capacity to identify a minimum core of predictors generally recognized as essential to pragmatically evaluate the risk of developing morbidity after heart surgery.
doi:10.1186/1472-6947-7-36
PMCID: PMC2222596  PMID: 18034873
7.  A comparative analysis of predictive models of morbidity in intensive care unit after cardiac surgery – Part I: model planning 
Background
Different methods have recently been proposed for predicting morbidity in intensive care units (ICU). The aim of the present study was to critically review a number of approaches for developing models capable of estimating the probability of morbidity in ICU after heart surgery. The study is divided into two parts. In this first part, popular models used to estimate the probability of class membership are grouped into distinct categories according to their underlying mathematical principles. Modelling techniques and intrinsic strengths and weaknesses of each model are analysed and discussed from a theoretical point of view, in consideration of clinical applications.
Methods
Models based on Bayes rule, k-nearest neighbour algorithm, logistic regression, scoring systems and artificial neural networks are investigated. Key issues for model design are described. The mathematical treatment of some aspects of model structure is also included for readers interested in developing models, though a full understanding of mathematical relationships is not necessary if the reader is only interested in perceiving the practical meaning of model assumptions, weaknesses and strengths from a user point of view.
Results
Scoring systems are very attractive due to their simplicity of use, although this may undermine their predictive capacity. Logistic regression models are trustworthy tools, although they suffer from the principal limitations of most regression procedures. Bayesian models seem to be a good compromise between complexity and predictive performance, but model recalibration is generally necessary. k-nearest neighbour may be a valid non parametric technique, though computational cost and the need for large data storage are major weaknesses of this approach. Artificial neural networks have intrinsic advantages with respect to common statistical models, though the training process may be problematical.
Conclusion
Knowledge of model assumptions and the theoretical strengths and weaknesses of different approaches are fundamental for designing models for estimating the probability of morbidity after heart surgery. However, a rational choice also requires evaluation and comparison of actual performances of locally-developed competitive models in the clinical scenario to obtain satisfactory agreement between local needs and model response. In the second part of this study the above predictive models will therefore be tested on real data acquired in a specialized ICU.
doi:10.1186/1472-6947-7-35
PMCID: PMC2212627  PMID: 18034872
8.  Estimation of progression of multi-state chronic disease using the Markov model and prevalence pool concept 
Background
We propose a simple new method for estimating progression of a chronic disease with multi-state properties by unifying the prevalence pool concept with the Markov process model.
Methods
Estimation of progression rates in the multi-state model is performed using the E-M algorithm. This approach is applied to data on Type 2 diabetes screening.
Results
Good convergence of estimations is demonstrated. In contrast to previous Markov models, the major advantage of our proposed method is that integrating the prevalence pool equation (that the numbers entering the prevalence pool is equal to the number leaving it) into the likelihood function not only simplifies the likelihood function but makes estimation of parameters stable.
Conclusion
This approach may be useful in quantifying the progression of a variety of chronic diseases.
doi:10.1186/1472-6947-7-34
PMCID: PMC2241590  PMID: 17996074
9.  A web-based laboratory information system to improve quality of care of tuberculosis patients in Peru: functional requirements, implementation and usage statistics 
Background
Multi-drug resistant tuberculosis patients in resource-poor settings experience large delays in starting appropriate treatment and may not be monitored appropriately due to an overburdened laboratory system, delays in communication of results, and missing or error-prone laboratory data. The objective of this paper is to describe an electronic laboratory information system implemented to alleviate these problems and its expanding use by the Peruvian public sector, as well as examine the broader issues of implementing such systems in resource-poor settings.
Methods
A web-based laboratory information system "e-Chasqui" has been designed and implemented in Peru to improve the timeliness and quality of laboratory data. It was deployed in the national TB laboratory, two regional laboratories and twelve pilot health centres. Using needs assessment and workflow analysis tools, e-Chasqui was designed to provide for improved patient care, increased quality control, and more efficient laboratory monitoring and reporting.
Results
Since its full implementation in March 2006, 29,944 smear microscopy, 31,797 culture and 7,675 drug susceptibility test results have been entered. Over 99% of these results have been viewed online by the health centres. High user satisfaction and heavy use have led to the expansion of e-Chasqui to additional institutions. In total, e-Chasqui will serve a network of institutions providing medical care for over 3.1 million people. The cost to maintain this system is approximately US$0.53 per sample or 1% of the National Peruvian TB program's 2006 budget.
Conclusion
Electronic laboratory information systems have a large potential to improve patient care and public health monitoring in resource-poor settings. Some of the challenges faced in these settings, such as lack of trained personnel, limited transportation, and large coverage areas, are obstacles that a well-designed system can overcome. e-Chasqui has the potential to provide a national TB laboratory network in Peru. Furthermore, the core functionality of e-Chasqui as been implemented in the open source medical record system OpenMRS for other countries to use.
doi:10.1186/1472-6947-7-33
PMCID: PMC2198908  PMID: 17963522
10.  Deafness mutation mining using regular expression based pattern matching 
Background
While keyword based queries of databases such as Pubmed are frequently of great utility, the ability to use regular expressions in place of a keyword can often improve the results output by such databases. Regular expressions can allow for the identification of element types that cannot be readily specified by a single keyword and can allow for different words with similar character sequences to be distinguished.
Results
A Perl based utility was developed to allow the use of regular expressions in Pubmed searches, thereby improving the accuracy of the searches.
Conclusion
This utility was then utilized to create a comprehensive listing of all DFN deafness mutations discussed in Pubmed records containing the keywords "human ear".
doi:10.1186/1472-6947-7-32
PMCID: PMC2180167  PMID: 17961241
11.  Is OpenSDE an alternative for dedicated medical research databases? An example in coronary surgery 
Background
When using a conventional relational database approach to collect and query data in the context of specific clinical studies, a study with a new data set usually requires the design of a new database and entry forms. OpenSDE (SDE = Structured Data Entry) is intended to provide a flexible and intuitive way to create databases and entry forms for the collection of data in a structured format.
This study illustrates the use of OpenSDE as a potential alternative to a conventional approach with respect to data modelling, database creation, data entry, and data extraction.
Methods
A database and entry forms are created using OpenSDE and MSAccess to support collection of coronary surgery data, based on the Adult Cardiac Surgery Data Set of the Society of Thoracic Surgeons. Data of 52 cases are entered and nine different queries are designed, and executed on both databases.
Results
Design of the data model and the creation of entry forms were experienced as more intuitive and less labor intensive with OpenSDE. Both resulting databases provided sufficient expressiveness to accommodate the data set. Data entry was more flexible with OpenSDE. Queries produced equal and correct results with comparable effort.
Conclusion
For prospective studies involving well-defined and straight forward data sets, OpenSDE deserves to be considered as an alternative to the conventional approach.
doi:10.1186/1472-6947-7-31
PMCID: PMC2173886  PMID: 17953759
12.  Instruments to assess the perception of physicians in the decision-making process of specific clinical encounters: a systematic review 
Background
The measurement of processes and outcomes that reflect the complexity of the decision-making process within specific clinical encounters is an important area of research to pursue. A systematic review was conducted to identify instruments that assess the perception physicians have of the decision-making process within specific clinical encounters.
Methods
For every year available up until April 2007, PubMed, PsycINFO, Current Contents, Dissertation Abstracts and Sociological Abstracts were searched for original studies in English or French. Reference lists from retrieved studies were also consulted. Studies were included if they reported a self-administered instrument evaluating physicians' perceptions of the decision-making process within specific clinical encounters, contained sufficient description to permit critical appraisal and presented quantitative results based on administering the instrument. Two individuals independently assessed the eligibility of the instruments and abstracted information on their conceptual underpinnings, main evaluation domain, development, format, reliability, validity and responsiveness. They also assessed the quality of the studies that reported on the development of the instruments with a modified version of STARD.
Results
Out of 3431 records identified and screened for evaluation, 26 potentially relevant instruments were assessed; 11 met the inclusion criteria. Five instruments were published before 1995. Among those published after 1995, five offered a corresponding patient version. Overall, the main evaluation domains were: satisfaction with the clinical encounter (n = 2), mutual understanding between health professional and patient (n = 2), mental workload (n = 1), frustration with the clinical encounter (n = 1), nurse-physician collaboration (n = 1), perceptions of communication competence (n = 2), degree of comfort with a decision (n = 1) and information on medication (n = 1). For most instruments (n = 10), some reliability and validity criteria were reported in French or English. Overall, the mean number of items on the modified version of STARD was 12.4 (range: 2 to 18).
Conclusion
This systematic review provides a critical appraisal and repository of instruments that assess the perception physicians have of the decision-making process within specific clinical encounters. More research is needed to pursue the validation of the existing instruments and the development of patient versions. This will help researchers capture the complexity of the decision-making process within specific clinical encounters.
doi:10.1186/1472-6947-7-30
PMCID: PMC2151936  PMID: 17937801
13.  Online detection and quantification of epidemics 
Background
Time series data are increasingly available in health care, especially for the purpose of disease surveillance. The analysis of such data has long used periodic regression models to detect outbreaks and estimate epidemic burdens. However, implementation of the method may be difficult due to lack of statistical expertise. No dedicated tool is available to perform and guide analyses.
Results
We developed an online computer application allowing analysis of epidemiologic time series. The system is available online at . The data is assumed to consist of a periodic baseline level and irregularly occurring epidemics. The program allows estimating the periodic baseline level and associated upper forecast limit. The latter defines a threshold for epidemic detection. The burden of an epidemic is defined as the cumulated signal in excess of the baseline estimate. The user is guided through the necessary choices for analysis. We illustrate the usage of the online epidemic analysis tool with two examples: the retrospective detection and quantification of excess pneumonia and influenza (P&I) mortality, and the prospective surveillance of gastrointestinal disease (diarrhoea).
Conclusion
The online application allows easy detection of special events in an epidemiologic time series and quantification of excess mortality/morbidity as a change from baseline. It should be a valuable tool for field and public health practitioners.
doi:10.1186/1472-6947-7-29
PMCID: PMC2151935  PMID: 17937786
14.  Modeling and detection of respiratory-related outbreak signatures 
Background
Time series methods are commonly used to detect disease outbreak signatures (e.g., signals due to influenza outbreaks and anthrax attacks) from varying respiratory-related diagnostic or syndromic data sources. Typically this involves two components: (i) Using time series methods to model the baseline background distribution (the time series process that is assumed to contain no outbreak signatures), (ii) Detecting outbreak signatures using filter-based time series methods.
Methods
We consider time series models for chest radiograph data obtained from Midwest children's emergency departments. These models incorporate available covariate information such as patient visit counts and smoothed ambient temperature series, as well as time series dependencies on daily and weekly seasonal scales. Respiratory-related outbreak signature detection is based on filtering the one-step-ahead prediction errors obtained from the time series models for the respiratory-complaint background.
Results
Using simulation experiments based on a stochastic model for an anthrax attack, we illustrate the effect of the choice of filter and the statistical models upon radiograph-attributed outbreak signature detection.
Conclusion
We demonstrate the importance of using seasonal autoregressive integrated average time series models (SARIMA) with covariates in the modeling of respiratory-related time series data. We find some homogeneity in the time series models for the respiratory-complaint backgrounds across the Midwest emergency departments studied. Our simulations show that the balance between specificity, sensitivity, and timeliness to detect an outbreak signature differs by the emergency department and the choice of filter. The linear and exponential filters provide a good balance.
doi:10.1186/1472-6947-7-28
PMCID: PMC2203979  PMID: 17919318
15.  Generating prior probabilities for classifiers of brain tumours using belief networks 
Background
Numerous methods for classifying brain tumours based on magnetic resonance spectra and imaging have been presented in the last 15 years. Generally, these methods use supervised machine learning to develop a classifier from a database of cases for which the diagnosis is already known. However, little has been published on developing classifiers based on mixed modalities, e.g. combining imaging information with spectroscopy. In this work a method of generating probabilities of tumour class from anatomical location is presented.
Methods
The method of "belief networks" is introduced as a means of generating probabilities that a tumour is any given type. The belief networks are constructed using a database of paediatric tumour cases consisting of data collected over five decades; the problems associated with using this data are discussed. To verify the usefulness of the networks, an application of the method is presented in which prior probabilities were generated and combined with a classification of tumours based solely on MRS data.
Results
Belief networks were constructed from a database of over 1300 cases. These can be used to generate a probability that a tumour is any given type. Networks are presented for astrocytoma grades I and II, astrocytoma grades III and IV, ependymoma, pineoblastoma, primitive neuroectodermal tumour (PNET), germinoma, medulloblastoma, craniopharyngioma and a group representing rare tumours, "other". Using the network to generate prior probabilities for classification improves the accuracy when compared with generating prior probabilities based on class prevalence.
Conclusion
Bayesian belief networks are a simple way of using discrete clinical information to generate probabilities usable in classification. The belief network method can be robust to incomplete datasets. Inclusion of a priori knowledge is an effective way of improving classification of brain tumours by non-invasive methods.
doi:10.1186/1472-6947-7-27
PMCID: PMC2040142  PMID: 17877822
16.  Developing a web-based information resource for palliative care: an action-research inspired approach 
Background
General Practitioners and community nurses rely on easily accessible, evidence-based online information to guide practice. To date, the methods that underpin the scoping of user-identified online information needs in palliative care have remained under-explored. This paper describes the benefits and challenges of a collaborative approach involving users and experts that informed the first stage of the development of a palliative care website [1].
Method
The action research-inspired methodology included a panel assessment of an existing palliative care website based in Victoria, Australia; a pre-development survey (n = 197) scoping potential audiences and palliative care information needs; working parties conducting a needs analysis about necessary information content for a redeveloped website targeting health professionals and caregivers/patients; an iterative evaluation process involving users and experts; as well as a final evaluation survey (n = 166).
Results
Involving users in the identification of content and links for a palliative care website is time-consuming and requires initial resources, strong networking skills and commitment. However, user participation provided crucial information that led to the widened the scope of the website audience and guided the development and testing of the website. The needs analysis underpinning the project suggests that palliative care peak bodies need to address three distinct audiences (clinicians, allied health professionals as well as patients and their caregivers).
Conclusion
Web developers should pay close attention to the content, language, and accessibility needs of these groups. Given the substantial cost associated with the maintenance of authoritative health information sites, the paper proposes a more collaborative development in which users can be engaged in the definition of content to ensure relevance and responsiveness, and to eliminate unnecessary detail. Access to volunteer networks forms an integral part of such an approach.
doi:10.1186/1472-6947-7-26
PMCID: PMC2194759  PMID: 17854509
17.  Indivo: a personally controlled health record for health information exchange and communication 
Background
Personally controlled health records (PCHRs), a subset of personal health records (PHRs), enable a patient to assemble, maintain and manage a secure copy of his or her medical data. Indivo (formerly PING) is an open source, open standards PCHR with an open application programming interface (API).
Results
We describe how the PCHR platform can provide standard building blocks for networked PHR applications. Indivo allows the ready integration of diverse sources of medical data under a patient's control through the use of standards-based communication protocols and APIs for connecting PCHRs to existing and future health information systems.
Conclusion
The strict and transparent personal control model is designed to encourage widespread participation by patients, healthcare providers and institutions, thus creating the ecosystem for development of innovative, consumer-focused healthcare applications.
doi:10.1186/1472-6947-7-25
PMCID: PMC2048946  PMID: 17850667
18.  Access, use and perceptions regarding Internet, cell phones and PDAs as a means for health promotion for people living with HIV in Peru 
Background
Internet tools, cell phones, and other information and communication technologies are being used by HIV-positive people on their own initiative. Little is known about the perceptions of HIV-positive people towards these technologies in Peru. The purpose of this paper is to report on perceptions towards use of information and communication technologies as a means to support antiretroviral medication adherence and HIV transmission risk reduction.
Methods
We conducted a qualitative study (in-depth interviews) among adult people living with HIV in two community-based clinics in Peru.
Results
31 HIV-positive individuals in Lima were interviewed (n = 28 men, 3 women). People living with HIV in Peru are using tools such as cell phones, and the Internet (via E-mail, chat, list-serves) to support their HIV care and to make social and sexual connections. In general, they have positive perceptions about using the Internet, cell phones and PDAs for HIV health promotion interventions.
Conclusion
Health promotion interventions using information and communication technology tools among people living with HIV in resource-constrained settings may be acceptable and feasible, and can build on existing patterns of use.
doi:10.1186/1472-6947-7-24
PMCID: PMC2048945  PMID: 17850656
19.  Clinical decision modeling system 
Background
Decision analysis techniques can be applied in complex situations involving uncertainty and the consideration of multiple objectives. Classical decision modeling techniques require elicitation of too many parameter estimates and their conditional (joint) probabilities, and have not therefore been applied to the problem of identifying high-performance, cost-effective combinations of clinical options for diagnosis or treatments where many of the objectives are unknown or even unspecified.
Methods
We designed a Java-based software resource, the Clinical Decision Modeling System (CDMS), to implement Naïve Decision Modeling, and provide a use case based on published performance evaluation measures of various strategies for breast and lung cancer detection. Because cost estimates for many of the newer methods are not yet available, we assume equal cost. Our use case reveals numerous potentially high-performance combinations of clinical options for the detection of breast and lung cancer.
Results
Naïve Decision Modeling is a highly practical applied strategy which guides investigators through the process of establishing evidence-based integrative translational clinical research priorities. CDMS is not designed for clinical decision support. Inputs include performance evaluation measures and costs of various clinical options. The software finds trees with expected emergent performance characteristics and average cost per patient that meet stated filtering criteria. Key to the utility of the software is sophisticated graphical elements, including a tree browser, a receiver-operator characteristic surface plot, and a histogram of expected average cost per patient. The analysis pinpoints the potentially most relevant pairs of clinical options ('critical pairs') for which empirical estimates of conditional dependence may be critical. The assumption of independence can be tested with retrospective studies prior to the initiation of clinical trials designed to estimate clinical impact. High-performance combinations of clinical options may exist for breast and lung cancer detection.
Conclusion
The software could be found useful in simplifying the objective-driven planning of complex integrative clinical studies without requiring a multi-attribute utility function, and it could lead to efficient integrative translational clinical study designs that move beyond simple pair wise competitive studies. Collaborators, who traditionally might compete to prioritize their own individual clinical options, can use the software as a common framework and guide to work together to produce increased understanding on the benefits of using alternative clinical combinations to affect strategic and cost-effective clinical workflows.
doi:10.1186/1472-6947-7-23
PMCID: PMC2131745  PMID: 17697328
20.  Group differences in physician responses to handheld presentation of clinical evidence: a verbal protocol analysis 
Background
To identify individual differences in physicians' needs for the presentation of evidence resources and preferences for mobile devices.
Methods
Within-groups analysis of responses to semi-structured interviews. Interviews consisted of using prototypes in response to task-based scenarios. The prototypes were implemented on two different form factors: a tablet style PC and a pocketPC. Participants were from three user groups: general internists, family physicians and medicine residents, and from two different settings: urban and semi-urban. Verbal protocol analysis, which consists of coding utterances, was conducted on the transcripts of the testing sessions. Statistical relationships were investigated between staff physicians' and residents' background variables, self-reported experiences with the interfaces, and verbal code frequencies.
Results
47 physicians were recruited from general internal medicine, family practice clinics and a residency training program. The mean age of participants was 42.6 years. Physician specialty had a greater effect on device and information-presentation preferences than gender, age, setting or previous technical experience. Family physicians preferred the screen size of the tablet computer and were less concerned about its portability. Residents liked the screen size of the tablet, but preferred the portability of the pocketPC. Internists liked the portability of the pocketPC, but saw less advantage to the large screen of the tablet computer (F[2,44] = 4.94, p = .012).
Conclusion
Different types of physicians have different needs and preferences for evidence-based resources and handheld devices. This study shows how user testing can be incorporated into the process of design to inform group-based customization.
doi:10.1186/1472-6947-7-22
PMCID: PMC1976086  PMID: 17655759
22.  Physicians' intentions and use of three patient decision aids 
Background
Decision aids are evidence based tools that assist patients in making informed values-based choices and supplement the patient-clinician interaction. While there is evidence to show that decision aids improve key indicators of patients' decision quality, relatively little is known about physicians' acceptance of decision aids or factors that influence their decision to use them. The purpose of this study was to describe physicians' perceptions of three decision aids, their expressed intent to use them, and their subsequent use of them.
Methods
We conducted a cross-sectional survey of random samples of Canadian respirologists, family physicians, and geriatricians. Three decision aids representing a range of health decisions were evaluated. The survey elicited physicians' opinions on the characteristics of the decision aid and their willingness to use it. Physicians who indicated a strong likelihood of using the decision aid were contacted three months later regarding their actual use of the decision aid.
Results
Of the 580 eligible physicians, 47% (n = 270) returned completed questionnaires. More than 85% of the respondents felt the decision aid was well developed and that it presented the essential information for decision making in an understandable, balanced, and unbiased manner. A majority of respondents (>80%) also felt that the decision aid would guide patients in a logical way, preparing them to participate in decision making and to reach a decision. Fewer physicians (<60%) felt the decision aid would improve the quality of patient visits or be easily implemented into practice and very few (27%) felt that the decision aid would save time. Physicians' intentions to use the decision aid were related to their comfort with offering it to patients, the decision aid topic, and the perceived ease of implementing it into practice. While 54% of the surveyed physicians indicated they would use the decision aid, less than a third followed through with this intention.
Conclusion
Despite strong support for the format, content, and quality of patient decision aids, and physicians' stated intentions to adopt them into clinical practice, most did not use them within three months of completing the survey. There is a wide gap between intention and behaviour. Further research is required to study the determinants of this intention-behaviour gap and to develop interventions aimed at barriers to physicians' use of decision aids.
doi:10.1186/1472-6947-7-20
PMCID: PMC1931587  PMID: 17617908
23.  Design of a Two-level Adaptive Multi-Agent System for Malaria Vectors driven by an ontology 
Background
The understanding of heterogeneities in disease transmission dynamics as far as malaria vectors are concerned is a big challenge. Many studies while tackling this problem don't find exact models to explain the malaria vectors propagation.
Methods
To solve the problem we define an Adaptive Multi-Agent System (AMAS) which has the property to be elastic and is a two-level system as well. This AMAS is a dynamic system where the two levels are linked by an Ontology which allows it to function as a reduced system and as an extended system. In a primary level, the AMAS comprises organization agents and in a secondary level, it is constituted of analysis agents. Its entry point, a User Interface Agent, can reproduce itself because it is given a minimum of background knowledge and it learns appropriate "behavior" from the user in the presence of ambiguous queries and from other agents of the AMAS in other situations.
Results
Some of the outputs of our system present a series of tables, diagrams showing some factors like Entomological parameters of malaria transmission, Percentages of malaria transmission per malaria vectors, Entomological inoculation rate. Many others parameters can be produced by the system depending on the inputted data.
Conclusion
Our approach is an intelligent one which differs from statistical approaches that are sometimes used in the field. This intelligent approach aligns itself with the distributed artificial intelligence. In terms of fight against malaria disease our system offers opportunities of reducing efforts of human resources who are not obliged to cover the entire territory while conducting surveys. Secondly the AMAS can determine the presence or the absence of malaria vectors even when specific data have not been collected in the geographical area. In the difference of a statistical technique, in our case the projection of the results in the field can sometimes appeared to be more general.
doi:10.1186/1472-6947-7-19
PMCID: PMC1925067  PMID: 17605778
24.  Improving search filter development: a study of palliative care literature 
Background
It is difficult to systematically search for literature relevant to palliative care in general medical journals. A previously developed search filter for use on OVID Medline validated using a gold standard set of references identified through hand searching, achieved an unacceptably low sensitivity (45.4%). Retrieving relevant literature is integral to support evidence based practice, and understanding the nature of the incorrectly excluded citations (false negatives) using the filter may lead to improvement in the filter's performance.
Methods
The objectives were to describe the nature of subjects reflected in the false negative citations and to empirically improve the sensitivity of the search filter. A thematic analysis of MeSH terms by three independent reviewers was used to describe the subject coverage of the missed records. Using a frequency analysis of MeSH terms, those headings which could individually contribute at least 2.5% to sensitivity (occurring 19 or more times) were added to the search filter. All previously run searches were rerun at the same time as the revised filter, and results compared.
Results
Thematic analysis of MeSH terms identified thirteen themes reflected in the missing records, none of them intrinsically palliative. The addition of six MeSH terms to the existing search filter (physician-patient relations, prognosis, quality of life, survival rate, treatment outcome and attitude to health) led to an increase in sensitivity from 46.3% to 64.7%, offset by a decrease in precision from 72.6% to 21.9%.
Conclusion
The filter's sensitivity was successfully increased using frequency analysis of MeSH terms, offset by a decrease in precision. A thematic analysis of MeSH terms for the false negative citations confirmed the absence of any intrinsically palliative theme or term, suggesting that future improvements to search filters for palliative care literature will first depend on better identifying how clinicians and researchers conceptualise palliative care. It is suggested that a constellation of parameters: stage of disease (advanced or active), prospect of cure (little or none), and treatment goals (primarily quality of life) may ultimately inform search strategies. This may be similarly true for chronic diseases, which share the inherent passage of time which marks them apart from acute, and therefore more readily identifiable, episodes of care.
doi:10.1186/1472-6947-7-18
PMCID: PMC1931586  PMID: 17597549
25.  An automatic method to generate domain-specific investigator networks using PubMed abstracts 
Background
Collaboration among investigators has become critical to scientific research. This includes ad hoc collaboration established through personal contacts as well as formal consortia established by funding agencies. Continued growth in online resources for scientific research and communication has promoted the development of highly networked research communities. Extending these networks globally requires identifying additional investigators in a given domain, profiling their research interests, and collecting current contact information. We present a novel strategy for building investigator networks dynamically and producing detailed investigator profiles using data available in PubMed abstracts.
Results
We developed a novel strategy to obtain detailed investigator information by automatically parsing the affiliation string in PubMed records. We illustrated the results by using a published literature database in human genome epidemiology (HuGE Pub Lit) as a test case. Our parsing strategy extracted country information from 92.1% of the affiliation strings in a random sample of PubMed records and in 97.0% of HuGE records, with accuracies of 94.0% and 91.0%, respectively. Institution information was parsed from 91.3% of the general PubMed records (accuracy 86.8%) and from 94.2% of HuGE PubMed records (accuracy 87.0). We demonstrated the application of our approach to dynamic creation of investigator networks by creating a prototype information system containing a large database of PubMed abstracts relevant to human genome epidemiology (HuGE Pub Lit), indexed using PubMed medical subject headings converted to Unified Medical Language System concepts. Our method was able to identify 70–90% of the investigators/collaborators in three different human genetics fields; it also successfully identified 9 of 10 genetics investigators within the PREBIC network, an existing preterm birth research network.
Conclusion
We successfully created a web-based prototype capable of creating domain-specific investigator networks based on an application that accurately generates detailed investigator profiles from PubMed abstracts combined with robust standard vocabularies. This approach could be used for other biomedical fields to efficiently establish domain-specific investigator networks.
doi:10.1186/1472-6947-7-17
PMCID: PMC1931433  PMID: 17584920

Results 1-25 (41)