First Place: Axial Exchange ($25,000)
Second Place: iBlueButton ($10,000)
Third Place:VoIDSPAN ($5,000)

Increasingly multidisciplinary solutions that are translated from one context to another have led to major successes in public health. Therefore, I see a huge research gap in many of the developing world countries that are being funded by the U.S. Global Health Initiative. I strongly believe that developing an online application that creates a global community of open-source data sources would be extremely useful to research. Many common digital aids, such as Google Maps, Zotero, and RSS feeds, could be leveraged to help researchers seek and present information more effectively. My idea for AcademyHealth REACH Challenge is to develop a platform that organizes pre-defined RSS feeds for different research disciplines. These RSS feeds would be tailored to each countries using all available data sources. U.S. data sources., particularly from the Nationwide Health Information Network, could be incorporated into this platform as an example for other countries to build. This idea would not directly impact clinicians and the public here in the US, but rather set up a system to translate lessons between countries supported by US funding, and thereby improve global health at large.

CONTACT THIS TEAM.

Our idea focuses on improving primary care provider (PCP) access to relevant, evidence-based information that supports clinical decisions and meaningful engagement for a population with chronic disease. A major barrier to improved outcomes involves a continued lack of an ability to access documentation that is required for assessment of Healthcare Effectiveness Data and Information Set (HEDIS) measure compliance – in this instance, a key component of high quality care required for comprehensive management of individuals with diabetes mellitus (DM). We recognize an opportunity to apply Natural Language Processing (NLP) technology to unstructured text through the EHR user interface to inform primary care providers (PCP) of appropriate documentation of eye examination requirements for improving compliance as set forth by HEDIS requirements for DM care. This could be accomplished via a provider-facing application that displays relevant information (e.g. alerts) within the EHR. Health information exchanges (HIEs) will need to be employed to aggregate and transmit the information across settings of care. This application will allow for better patient-centered care and improved population health by impacting the uptake and documentation of recommended diabetes-related eye care, a key recommendation in the HEDIS comprehensive diabetes care measure set.

There are persistent gaps in several areas related to our identified issue. A recent systematic review revealed that there is very little evidence of any DM-related advanced HIT functionalities or HIE-related applications specifically within eye and vision care directed at the multi-disciplinary care issues surrounding the collection of documentation of eye-related care for individuals with chronic DM that both 1) supports patient and population health and 2) assists in improving quality care delivery and process outcomes. For example, important quality data exist regarding the referral by other health care providers for, and uptake by patients of, eye examination for individuals with DM as per the HEDIS measure set recommendations. Recent state-specific HEDIS rates for DM-related eye/retinal examination indicated that on average only 58% of individuals with DM receive documented timely retinal eye exams and show little improvement year to year, and that PCP medical records analysis indicated that 76% of patient records had no documentation of meeting requirements for needed care. As a result, there is a lack of consensus on how to best guide the efforts of eye care and primary care clinicians and individuals with DM towards the intervention – or combination of interventions – that can improve upon these findings, and a gap in knowledge-driven decision support remains in how to use HIT to prevent delays in treatment, effectively reduce the risk for chronic vision impairment, and avert poor health-related quality of life outcomes in a population requiring specific documentation guidelines for quality measures involving a multi-disciplinary health care team. One way to overcome this challenge is to facilitate information access from the HIE via an NLP application that assists in uptake of eye examination data available from all HIE inputs, especially from those outside of the PCP-managed EHR.

Another gap involves the merger of information from outside the PCP setting (such as that contained within electronic patient records or EPR) with what is, and can be, maintained in the EHR. As health data contained in the EPR is patient controlled, curated primarily by the individual patient, and often contains information from multiple contributors regarding the individual’s health, there are barriers to pertinent information finding its way into the EHR and becoming part of the legal patient record. Although EPR may hold great potential value for improving population health via their use in the management of chronic disease, the current evidence for this remains sparse. Further, even though most patient data entry systems have been created independent of EHR, EPR have been found to be closely linked with clinician EHR systems and as such, their adoption continues to be dependent upon growth in EHR adoption. An application that can bridge these issues and utilize all available health information via HIEs would be of great value for clinicians managing chronic disease and for the health care organizations that must monitor and report on quality of care. It would also allow for information to be drawn from stakeholders in a broad range of environments including acute and chronic inpatient care facilities, an array of ambulatory care settings, community health centers, and even within patient homes.

Other broader challenges exist for HIT systems relative to the unique workflow and data management issues of chronic disease management care providers due to data management issues that require higher-level functionalities beyond the EHR “basics”. For instance, only recently has there been consensus on recommendations used to inform healthcare standards development organizations (SDO) regarding EHR systems and their employ in meeting the needs of eye care consumers in the U.S. in outpatient settings (HL7 eye care group). As a result, the identification of unique functions and specific conformance criteria critical for inclusion in EHR systems through which a patient might receive eye and vision care are only beginning to be identified and adopted. Translating DM eye care information into the PCPs primary EHR is a necessary part of the process, and NLP can be applied to assist in filling this gap.

Our aim is to develop a patient and eye-care provider-accessible application employing NLP technology for use in HIEs like NHIN Direct to inform primary care providers (PCP) of appropriate documentation of eye examination requirements for improving compliance and meeting HEDIS requirements for documentation of diabetes-related eye care. The audience includes patients, eye care providers and primary care providers , and additional end-users like managed care organizations, population health professionals, NCQA, etc. We desire to gather relevant health care information and use data from Structured and unstructured text in electronic health records (EHR), personal health records (EPR), and other information systems for providers related to HEDIS requirements (NCQA). Also, state-specific HEDIS data can be used to measure outcomes of this project.

We desire to focus initially on the specific function of gathering relevant health care information documentation from the HIE (e.g. NHIN Direct) and employ NLP technology to translate care requirements within unstructured data into actionable data. By using HIEs, pertinent health information can be referenced and collected across multiple care settings, regardless of which information system is being used. Through NLP technology, the bulk of provider documentation, which is being stored as free-text notes within an electronic system, can be analyzed, contextually translated, and indexed such that the information can be utilized in discrete ways, including through alerts and data mining/analytics. Using these two technologies, the necessary documentation pieces from various health care settings can be accessed and analyzed for completion of HEDIS measures. As a result, alerts and other reporting can be employed to prompt providers to complete any omitted HEDIS measures and/or follow up with the appropriate provider to ensure the patients are receiving appropriate and comprehensive care.

At this time, we bring several strengths to the project. In addition to clinical expertise in diabetic eye disease and related content areas (e.g. CMS policy) including clinical management of chronic vision impairment related to DM, we bring a population-based perspective from comprehensive informatics education through a new formal certificate training program recently completed in the Department of Health Services Research at the Johns Hopkins Bloomberg School of Public Health. Additionally, we bring direct experience from working with Maryland’s state-designated Health Information Exchange “CRISP”, and most recently through experience with a machine-learning natural language processing medical coding software company. Lastly, we bring passion to this project from both our perspectives for employing a HIT interdisciplinary team approach to meet the stated objective. We seek to gain the expertise of other team members in the areas of development, including programmers and other IT specialists skilled in NLP and HIEs who share our passion to improve the health and visual outcomes of this patient population.

CONTACT THIS TEAM.

Therefore, I propose building an application for people with diabetes and heart disease — two of he most common chronic conditions — that presents retail pricing information for commonly-used products and services, based on the publicly-available Medicare reimbursement fee schedule. Health care costs are usually based on grossly-inflated, never-actually-reimbursed charges. Let us open up and make convenient information about the actual paid reimbursement for health care products and services!

To this project I can contribute research experience as a health services researcher and technical information about health care product and service pricing. In the past I have led a couple software development projects, which will me be a better collaborator with coders and engineering professionals.

CONTACT THIS TEAM.

The already existing “childhealthdata” website is a resource provided through the CAHMI led Data Resource Center for Child and Adolescent Health (DRC), which was developed and is supported through a cooperative agreement between the CAHMI and the federal Maternal and Child Health Bureau (MCHB), Health Resources and Services Administration, U.S. Dept of Health & Human Services). The proposed “childhealthdata” mobile application will allow users quick access to the most essential facets of these large datasets: (1) fast track access to state level prevalence data on over 200 child health, well-being and system performance indicators, easily stratified by race/ethnicity, socioeconomic status, type and adequacy of health insurance and child health status (including children with special health care needs); (2) interactive and in depth point and click querying on specific topics; (3) executive-level micro and macro summary data in graphical, tabular and narrative formats; and (4) connecting to the knowledge and experience of the CAHMI staff for technical assistance.

Audience:
Key audiences for the proposed “childhealthdata” mobile application are the same as for the existing DRC website, with hopes to further real-time use by health care policymakers, program leaders, journalists and media professionals and advocates. We also hope the mobile application will expand use by health care providers, quality measurement/improvement professionals, researchers and students who will be able to access data and information in contexts where full screen internet access is not convenient. Data accessible through the “childhealthdata” mobile application will be useful for informing decisions requiring valid and standardized data on prevalence and disparities in the health and health system performance for children, youth and families. It is common for real time assumptions about these topics to be inaccurate, whether it is about prevalence of health needs, gaps in health services, associations between demographic factors and health risks and so on. The proposed applications will be marketed as a tool to check assumptions underlying policy, program and research decisions during this time of heightened focus on children’s health brought about by CHIPRA and ACA legislation with a tagline such as “What have you learned about child health today?”.

Resources held and needed:
Currently held. The DRC Cooperative agreement has allowed the CAHMI to operate a set of relational databases (SQL Server 2008 R2) containing survey data, survey data analysis, and supplemental materials (e.g., documents). These databases are the primary tools that translate raw data from the NSCH and NS-CSHCN flat-file datasets into accessible and succinct portions of information. We have worked with the MCHB and CDC in developing content for the two surveys, and we have access to an executive level board of advisors who can help advise us on the project. We have some junior-level skill sets for mobile development, graphic design and have been using the Corona SDK in our development on the project to date, although we are not committed to this development kit.
Needed. The CAHMI is requesting additional developer staffing so that we can implement our design specifications for the “childhealthdata” App. In particular, a senior-level developer is needed to make key design decisions, and support our current software design capacity for the ‘heavy-lifting’ software engineering of the “childhealthdata” App system.

Why we are the ones:
The CAHMI staff have been active stewards of NSCH and NS-CSHCN data for more than a decade, and have worked with HRSA/MCHB and CDC/NCHS to develop many national indicator measures used to inform policy decisions and educate the public. We have a history of collaborating with healthcare providers, improvement partnerships, academic departments, and government agencies. Also, the CAHMI have been responsible for disseminating findings from the two national survey datasets since 2003, and we have an understanding of users’ needs regarding access to child health data.

Key development issues:
Content Selection & “Look and Feel”: Developing the proposed App will require a strategic paring-down of the content currently available on the existing “childhealthdata” website. This includes determining streamlined visual information for data queries, and selectively editing other functionalities such as state profiles and maps.
Additional Database Programming: We will need to augment our existing relational databases with new database objects (e.g., tables and queries).
IT Infrastructure: We will need to set up a Web service to translate data from we will need to set up an efficient way for the application to access the Web service.

Development status:
We have created a high-level design document for the proposed App, and the application is currently in a preliminary design stage. We will continue with the conceptual development of this App with input from the key audiences pending additional support for the technical and graphical build of a “childhealthdata” App.

CONTACT THIS TEAM.

Idea: Adherence to a physician-prescribed management program, including exercise, test scheduling, and insulin regulation, could improve the health condition of those living with this illness. This application seeks to improve patient activation that improves adherence to a physician-prescribed management program resulting in better management of diabetes within a select population though smartphone technology.

Audience: Those who are diabetic and taking insulin, overweight (25 kg/m2 to 39 kg/m2), aging (65+), and enrolled with a Primary Care Management Team in a diabetes registry.

Background: Currently, several modalities are used to “activate” patients (e.g., mail, phone, office visit). Such a system, Patient Engagement System (PES), already exists and is in use today.(3) The advent of Smartphone technologies increases opportunities to assist patients with self-managing their illnesses (e.g., text messages, emails, social networking, etc.). With over 230 million Americans using smartphone technologies, it has quickly become the communication channel of choice.(4) Free smartphone applications, including Glucose Buddy, Diabetes Buddy and WaveSense Diabetes (via iPhone) are data storage for various utilities. Track3-Diabetes Planner and Carb counter is also offered for a price of $5.99. However, these technologies also require the patient to input a tremendous amount of information. Too much information can be tedious for elders not accustomed to using all features of smartphones. Further, although email capabilities are offered, none offer an automatic update to patient providers.

Function: As a trial, the smartphone system will be delivered to private practices through a chosen system (e.g. Aprima HER). After construction of the application with the chosen system, it can be downloaded (free of charge) while at the physician’s office. The patient would agree to all HIPAA consents that their data would then be transmitted back to the patient registry on a system that the provider team (case manager and physician) can access. The information contained in the system would automatically update those who are also part of the shared documents (e.g., adult children, parents, nursing home personnel, or physician) appointed by the patient, when the patient has successfully completed any necessary patient activation activity.
The system would either call or text simple questions, requiring the patient to press a key to answer. Examples include:

Have you eaten today? Press or say 1 for Yes or 2 for No
Have you taken your insulin today? Press or say 1 for Yes or 2 for No
What time did you take your insulin: (could be one or all of these)
• This morning? (input answer)
• This afternoon? (input answer)
• This evening? ( input answer)

The answers are immediately updated in the system and made available to the case manager or physician. If a person is not able to answer the questions for one reason or another, the system is alerted and then further action (e.g. the office calls) can be taken.
The smartphone system allows for a sense of autonomy and security both at the same time.

CONTACT THIS TEAM

References
1 US Department of Health and Human Services. (2010). African American Women and Diabetes. Retrieved from http://www.womenshealth.gov/minority-health/african-americans/diabetes.cfm
2 Schoenberg, N., Traywick, L., Jacobs-Lawson, J. & Kart, C. (2008). Diabetes Self-care among a Multiethnic Sample of Older Adults. Journal of Cross Cultural Gerontology, 23, 361–376. doi 10.1007/s10823-008-9060-z
3The Patient Engagement System (PES). Retrieved fromhttp://vermedx.com/research.htm
4 Practical Ecommerence Staff (2010, April 6). Chart of the Week: Smartphone Usage Grows, Especially Android. Retrieved fromhttp://www.practicalecommerce.com/articles/1793-Chart-of-the-Week-Smartphone-Usage-Grows-Especially-Android

Functional Requirements:

1. Centralised Database containing nationwide registrant’s information.

2. Parameters taken from the users would be simple & registration shall be in a non obtrusive format.

3. Automated query & response.

Functionally:
a. If a person needs blood he shall just log in through web / SMS / call up helpdesk.
b. His request will be fed in the system automatically / manually in a pre-defined format.
c. The system will logically identify the locality to short-list potential donors in the vicinity.
d. Matching predefined parameters from the request (e.g. blood type) & the donors.
e. Automatic SMS will be sent to the result set & donors will be requested to reach out.
f. Respondents will be logged into the database & a response report will be generated. This report will be sent back to the user & the interested donors will be contacted.
g. Once some donor arrives & blood is donated, a confirmation SMS is sent by the hospital. This will trigger an automated SMS to all the people who have been sent the message in the first place as “One life saved”
h. Donor shall be eligible for some reward points (or any other incentive program).

Promotional Strategy:

4. Registration benefits include:
a. Free health checkups for all those who register & donate blood atleast once.
b. Social Service Certificate in alliance with affiliated NGOs.
c. Promotional bands or some gift when a person registers through a post office, an associated NGO or an awareness camp.

5. Web portal with general information & relevant articles in all major languages. This would be rich content to promote portal as an informative site.

6. Campaigns launched with the help of NGOs, Post offices & Local govt. bodies to have people register. This would not require much effort as we would only need a laptop & a wireless internet connection.

7. Medical benefits with voluntary support from doctors.

Technical Expertise / Resource Requirement:

I have extensively worked on database integration & query modelling. It’s helpful in creating a system that is robust & logically functional there by replicating the above business logic into a suitable IT model. I would be requiring a person who is proficient in developing IT design pattern & system sustainability. His expertise must include thorough knowledge of software life cycle & implementation. Front/ Back end support. Moreover, a person with experience in web designing would be an added advantage. He will monitor & enhance the web interface & consumer response. I have expertise in online marketing & analytics and this will help us in recovering the operating cost.

Dr (Major.) Manu Chopra, India will assist us on any information or help on medical grounds.

Return on Investment:

1. Annual Fee from Medical Operatives (Hospitals, Health Centers & Clinics)
2. Donations from various trusts and MNCs
3. Revenue from site operations
a. Article marketing
b. Advertisement revenues
c. Viral product promotions
d. Partnerships & sponsorships from medical organisations
e. Active doctor-patient discussions, blogs

CONTACT THIS TEAM

The Target Audience:
Given the identified gaps reflect the key link between provider organization and patient / families, the target audience for our project’s focused idea are healthcare organizations and the individuals who receive care. We believe this focused idea is highly accessible to the average healthcare consumer; those with smart phones will be able to access and use this technology. Currently 83% of Americans have a cell phone and 42% of those use smart phones, the fastest growing segment of cell phone users (Pew Research, 2011).

Focused Improvement Idea:
Develop hospital specific Quick Response (QR) codes in inpatient and outpatient service areas for patients/customers who would like to give structured and valuable feedback about their care experience at the time service is rendered. The data will be real-time and allow the provider to quickly react for service recovery or caregiver recognition; it will not contain medical information that would reveal any personal health information (inputting MRN/SSN information is optional). The QR code will link to an organization specific URL for customers to enter the following information:

Feedback Category:
- Clinical Care
- Communication
- Employee Behavior / Conduct
- Facility / Environment of Care
- Other (defined or left blank by customer)

Data to be Used:
Patient response data will be entered through the hospital-specific URL and stored securely in an internal database. It will be centrally managed by the organization allowing them to immediately alert care areas, providers, and ancillary services for prompt service recovery. Quick response time will facilitate increased provider/patient communication and reduce errors related to care provision. Positive feedback from patients can serve a valuable purpose for increasing workforce and department engagement, another priority focus for the healthcare industry. In the same fashion, the positive feedback will create public recognition for the quality of care delivered. Healthcare organizations will not only have the ability to internally recognize caregivers for their service, but can also use their collected feedback data to market themselves externally. Positive comments and service recovery data may be a useful marketing tool for describing a healthcare organization’s quality of care and compassion.

Our Team’s Skills and Knowledge:
Our team has completed a comprehensive literature review on the use of social media in healthcare. We have knowledge and skills associated with healthcare’s perception and use of social media, and the influence of social media on a healthcare organization’s reputation. We have training in translational research and understand how to bridge the gap between scientific evidence and practitioner/end user application of knowledge. Finally, our group is composed of trained healthcare professionals, clinical and non-clinical, healthcare researchers, and healthcare administrators. Our multi-disciplinary team composition contributes to our team’s greatest strength – diversity in theoretical foundations and varying roles related to the patient care experience, which greatly contributes to our being able to understand the need for and application of this technology.

Needed Skills and Knowledge:
We are seeking technology skills to assist in the development of: QR codes, websites, databases, RSS feeds, and alerts. We would need to build in the option for patients to input their MRN or SSN when entering feedback; we would also need an internal database created to collect the feedback, and a website for external use to market positive feedback.

CONTACT THIS TEAM.

1. Search and create a large set of literature relating to the specific disease
2. Extracts and categorize literature by like concepts into hierarchically organized folders
3. Transforms a seemingly complex set of literature into a hierarchy of common themes and displays them in a visual two dimensional landscape map for review

So researcher can spot predominant concepts, pinpoint areas of knowledge gaps, and quickly obtain the information you need to ask most important research questions

This application will not benefit individual patient in short-run, but I expect it will substantially benefit patient care at population level in long-term. In addition, literature mapping can also be applied in other scientific fields to identify hot spots and knowledge gaps.

CONTACT THIS TEAM