📕Background
Why did we develop Simple? How can a data system benefit a hypertension control program?
Last updated
Why did we develop Simple? How can a data system benefit a hypertension control program?
Last updated
Hypertension is the leading preventable cause of premature death worldwide, accounting for 10.7 million deaths per year, with most deaths occurring in low- and middle-income countries.
Large-scale hypertension control programs in low-resource settings face many challenges, including understaffed health systems with overworked staff often too busy to document detailed clinical data during patient encounters.
A strong public health program backed by a well-designed digital health information system has immense potential to save lives. An effective digital system must accommodate healthcare worker time constraints so the program can focus on providing high-quality clinical care.
A practical digital system is most likely to succeed if it: 1) is very fast and easy to use, 2) includes only a few key indicators, 3) requires minimal data entry, and 4) is designed with offline-first capability.
The best system is ultimately one that helps the most patients reduce their blood pressure.
Hypertension kills more people than all infectious diseases combined. Hypertension is also very common, affecting 20-35% of adults in most communities. That means that about 1/5 of adult patients in a hospital's outpatient department will present with hypertension. Practically, this means that enormous numbers of patients require counseling, measurement, and treatment — and those patients need to be monitored over time to ensure that their hypertension is controlled successfully.
As you can imagine, this volume of patient management is challenging in busy hospitals in places like India, Bangladesh, Sri Lanka, and Ethiopia.
17.9 million people die each year - from cardiovascular diseases (CVDs), an estimated 31% of all deaths worldwide (Statistics from 2017)
Out of 17 million premature deaths (under the age of 70) due to non-communicable diseases in 2015, 82% are in low and middle-income countries, and 37% are caused by CVDs. (source)
75% of CVD deaths occur in low and middle-income countries. (Statistics from 2017)
85% of all CVD deaths are due to heart attacks and strokes. (Statistics from 2017)
People with cardiovascular disease or who are at high cardiovascular risk due to the presence of one or more risk factors (such as hypertension, diabetes, hyperlipidaemia, or already established disease) need early detection and management using counseling and medicines, as appropriate. (source)
A system for monitoring is one of the core pillars of a hypertension control program based on the World Health Organization's HEARTS technical package. By recording a high volume of follow-up visits by patients with hypertension, it's possible improve the health system more quickly. Fundamentally, Simple is a system of feedback loops to improve patient treatment and strengthen health systems.
Just with minimal information about each patient's current BP measures, medications, a health system or hospital can monitor:
BP controlled
How many patients have visited recently with their blood pressure under 140/90? This is the critical indicator to measure a successful hypertension control program.
BP uncontrolled
How many patients visited recently but their blood pressure is not controlled yet? These patients need to be treated successfully to bring their BP under control.
Missed visits
How many patients are not regularly receiving care? These patients need to be encouraged to return to care.
Registrations
How many patients are enrolled in the hypertension program? How much of the estimated hypertension patients in the population does this represent?
In the graphs above, an epidemiologist or health official can read a story. BP control rates in region are increasing over the last 3 months, great! This is a result of declining "Missed visit" rates (blue chart) and also that more patients who come back have their BP controlled now (red chart). If we want to continue to increase BP control, we could focus on returning even more patients to care or we could try to reduce the red number further (e.g. by counseling patients on medication adherence, by ensuring medical officers are treating to protocol, or by sending patients home with longer prescriptions). This region has a good BP control rate (above 50% is good!), so maybe we would focus on identifying more people in the community with hypertension and start treating them — 35,525 might only represent 5% of the local population with hypertension.
Other indicators are also tracked. For instance:
Lost to follow-up patients: How many patients have not visited in over 1 year?
Calls made to patients: How many calls have healthcare workers made to patients?
Medication titration rates: How often are medical officers intensifying treatment for patients with uncontrolled blood pressure?
See our What we report section for a detailed list of indicators and definitions.
Patients want to know whether treatment is succeeding. Clinicians strive for better management of each patient. Health system managers and public health officers are concerned with system performance and entire populations, often millions of people.
The core challenge, therefore, is to design software that meets the needs of all three of these groups while operating within severe time constraints at the point of care.
Relationship to software
Uses individual blood pressure measurement data
Uses software to ensure each patient receives correct treatment
Uses aggregate data from software to determine system performance
Ultimate goal
Monitor their own progress toward blood pressure control
Monitor each patient’s progress to control blood pressure
Monitor blood pressure control of patient population
Critical needs
Monitor progress, visit convenient clinics
Quick overview of patient’s recent history
Big picture view of where blood pressure is controlled and where to focus effort
Constraints
Time constraints; hypertension treatment is a low priority
Roughly 15 seconds available for data entry; high turnover so easy training is key
Manages other programs; little time
Simple is designed to be a pragmatic tool. A typical clinical visit in India lasts 3-4 minutes. In Bangladesh it's closer to 2 minutes. A busy clinician is juggling many tasks in this tight time: taking blood pressure and blood sugar measures, diagnosising, prescribing, and counseling the patient.
Keep in mind that clinicians don't come to work to do data entry. They focus on treating patients.
If we expect thousands of overworked healthcare workers to enter a high volume of data for a large percentage of their patients, we need to make fast and easy.
A digital health information system that minimizes and streamlines data entry allows healthcare workers to concentrate their limited time on providing direct patient care. Registration of new patients requires entry of patient information, demographic data, and health history, which needs to be done only during initial intake. A well-designed digital system can accomplish patient registration and first clinical visit in approximately 80 seconds (exclusive of physical blood pressure measurement): scan ID card to determine whether the patient is already registered and if not decide whether to enroll as a new patient (6 seconds); verify or enter patient information and demographics and conduct brief survey of cardiovascular history and risk, including current medications (66 seconds); enter blood pressure readings (6 seconds); and schedule follow-up visit (2 seconds).
After initial registration, healthcare workers need only scan the patient ID card and verify identity, enter blood pressure readings, verify current medications, and schedule follow-up, all of which can be done reliably within about 15 seconds. Variables required for patient registration must also be kept to an absolute minimum or many patients will go unregistered and their data unrecorded. There are many important variables of interest to epidemiologists and program managers. By evaluating these additional variables in specially designed studies on specific groups rather than for every patient, more reliable data will be collected and, most importantly, front-line health workers’ time to interact with patients will be respected and protected.
Designing with the pretense that healthcare workers have more time than they do results in software that is never adopted, is used inconsistently, or collects inaccurate data. Usability by front-line staff should be prioritized over any other consideration.
A tool that is fast to use can also be fast to train; minimizing data entry minimizes what healthcare workers need to learn. Train users in situ at hospitals and clinics, ideally in less than 1 hour, to reduce overhead costs for conducting trainings and to enable implementation scale.
Staff turnover and task sharing are common. If the digital tool is easy to learn, trained staff can teach others.
Based on metadata from more than 1.8 million patients, the Simple mobile app is widely used and performance remains very fast (approximately 80 seconds for registration and entering data from the initial clinical visit and 15 seconds to enter follow-up visit data). Data dashboards are easy to produce and disseminate and widely used by program staff. In Bangladesh, hypertension control rates more than doubled within a few months, from approximately 20% to 45%, in clinics that adopted the Simple app, although other program improvements likely also contributed to this increase.
In qualitative interviews, healthcare workers consistently rate the Simple app highly and emphasize how it reduces their burden of work and helps them to efficiently manage patients, monitor progress to improve blood pressure control, and deliver better care to patients
The Resolve to Save Lives hypertension control program
See an example hypertension treatment protocol from Punjab, India
See "How to take a BP: BP Measurement checklist" to learn to take a BP
