View source for A Clinical Decision Support System for Integrating Tuberculosis and HIV Care in Kenya: A Human-Centered Design Approach

==Abstract==
With the aim of integrating HIV and tuberculosis care in rural Kenya, a team of researchers, clinicians, and technologists used the human-centered design approach to facilitate design, development, and deployment processes of new patient-specific TB clinical decision support system for medical providers. In Kenya, approximately 1.6 million people are living with HIV and have a 20-times higher risk of dying of tuberculosis. Although tuberculosis prevention and treatment medication is widely available, proven to save lives, and prioritized by the World Health Organization, ensuring that it reaches the most vulnerable communities remains challenging. Human-centered design, used in the fields of industrial design and information technology for decades, is an approach to improving the effectiveness and impact of innovations that has been scarcely used in the health field. Using this approach, our team followed a 3-step process, involving mixed methods assessment to (1) understand the situation through the collection and analysis of site observation sessions and key informant interviews; (2) develop a new [[CDS| clinical decision support system]] through iterative prototyping, end-user engagement, and usability testing; and, (3) implement and evaluate the system across 24 clinics in rural West Kenya. Through the application of this approach, we found that human-centered design facilitated the process of digital innovation in a complex and resource-constrained context.<ref>http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4149343/pdf/pone.0103205.pdf Caricia Caralani, 2014</ref>

==Summary==
===Introduction===
====HIV and Tuberculosis in Kenya====
The connection between HIV and TB is substantial, as people living with HIV have a 20 fold higher risk of dying from TB than those who are HIV-negative. Thus, TB in HIV-positive people has become a national health priority in Kenya.  Current literature suggests isoniazid preventative therapy (IPT) reduces overall TB incidences, but even with the priority set on TB and HIV, the rate of TB infections and TB-related deaths continues to rise.  To fix this, HIV care systems will need to become more effective at triaging and monitoring HIV patients.  AMPATH, one of the first clinical care networks used in Africa, has began to be improved by the TB Tech Project in January 2013.  Through the human-centered design (HCD) process, an innovative approach to improving the percentage of eligible HIV patients starting IPT was explored, refined, and delivered.
====Human-Centered Design====
HCD standards include the following:
*The design is driven and refined by HC evaluation
*The process is iterative
*The design addresses the whole user experience
*The design team includes multidisciplinary skills and perspectives.
While HCD is not explicitly a research method, it incorporates other methods to achieve the design goals.  Ultimately the aim of this approach is to enhance the usefulness, usability, and use of health technology so that health outcomes can be improved. 
====HCD for TB/HIV Technology in Kenya====
TB Tech used the IDEO approach for HCD, this involved three stages:
*Hear: researchers use to learn to understand social context and inspire new solutions for HIV and TB care, barriers to IPT initiation, EHRs and other data sources, and what influences a physicians care practices.
*Create: over a 6 month period collaborations translated the insights derived in the hear stage in the design of a TB clinical decision support system.
*Deliver: A final product was delivered as of March 2014 and by following the IDEO model further investigations will be made.
It is the goal of this study to explore the problems of TB/HIV and design a system to help improve the percentage of eligible HIV patients starting IPT.  And to quantify this goal the TB Tech team is currently leading an impact evaluation to measure the effect on the integration TB/HIV care.

===Methods===
====Data Collection====
=====Site Observations=====
Among the 49 active AMPATH sites 9 were selected for use.  They were selected for variability of rural/urban, average monthly patient volume, total number of active providers, and AMPATH's clinical categorization of 1-6 (1 being a village kiosk and 6 being a district referral hospital).  Site observations were held for 3-4 hours.
=====Key Informant Interview=====
Key informants were also selected for variability, the categories were health areas of expertise, role at AMPATH, department, and geographical responsibility.  Interviews of these informants lasted 45mins to an hour.  
=====Lab Simulation Testing=====
Using 217 pseudo-patients, the programming of the clinical decision support system that would be displayed during a patient visited was identified.  The lab simulation was run until all 217 patients received the appropriate decision support message.
=====In-Context Usability Testing=====
Usability testing occurred at 3 clinical sites among 10 HIV clinicians.  A mixed methods survey was then used to access the clinicians perceptions of the understandability, importance, helpfulness, practicality, and accuracy of each TB reminder message.  There was also an in-depth interview at the end of the day to provide feedback lasting 20-30 mins each.  
====Data Analysis====
Modified ground theory was used to analyze all data.

===Findings===
====Hear====
=====Clinician attitudes about IPT=====
*Difficult to manage TB priorities among other priorities
*providers remained hesitant to start a patient on IPT due to pharmacy shortages
=====Clinician knowledge of IPT=====
*most clinicians had insufficient knowledge to determine eligibility of patients 
=====Clinician perceptions of information systems=====
*happy about EMRs and in particular the HIV clinical decision support summary
*suffer from missing patient data
=====Clinical resources for determining IPT eligibility=====
*Chest x-rays were not available to small communities
*Lab tests were available to every site
====Create====
=====Lab simulation=====
*dozens of cycles were conducted until zero errors were reached
=====Usability Testing in context=====
*Well received receiving scores on a range of 4.2-4.5 on a scale of 5
*Many errors occurred approximately 1 in 4
*Almost 1 in 2 were unactionable
====Deliver====
*6 months spent on feasibility challenges

===Discussion===
Within the design there are two major critiques to HCD. One, due to the strong interaction with specific communities it is believed that the product may not work on a larger scale. And while this may end up being true, overly generalized health information technology has never been proved effective in any context.  Two, heavily relying on user feedback from users who may not know exactly what they want or need. It is important that key stakeholders take part in the design because if they feel as though they can't use it, then the whole design is essentially useless.

===Conclusion===
TB Tech successfully used HCD to construct a digital innovation in a resource constrained setting.  And the system was then implemented over 71 sites.

===Comments===
Interesting article, don't know how much stock I put into the system. If you look at the discussion section there are two very valid concerns with this paper's methods.  I also would like to see the stats on how well the implementation actually worked and what the overall patient outcomes were.

==References==
<references/>
[[Category: Reviews]]
[[Category: CDSS]]
[[Category:HI5313-2015-FALL]]