How many digital health conferences start with the same tired promise: “We just need to digitalize everything and health data quality will magically improve.” After reviewing PATH’s comprehensive assessment of primary health care data systems across Burkina Faso, Ethiopia, Kenya, and Nigeria, I can tell you that this Silicon Valley-inspired narrative is dangerously wrong.
The truth is more painful. Your data quality problems probably have nothing to do with lacking the right mobile application or data collection tool. They’re happening because:
- You are collecting more data than you use,
- Performance incentives are driving “data cooking,”
- Unsexy problems need funding more than innovations,
- Parallel systems are draining limited resources
- Simple solutions beat artificial intelligence
- Interoperability is possible with incentives
- Asking women health workers to do unpaid data entry while men attend the capacity-building workshops.
1. You’re Collecting Data You’ll Never Use
PATH documented something embarrassing: health systems across Africa are drowning in data they collect but never analyze. Community health workers fill out 20+ different registers. Facilities compile monthly reports that districts validate and enter into national systems. Then the data sits there, unexamined.
The fix isn’t another dashboard.
It’s the unglamorous work of identifying which indicators actually inform decisions and stopping collection of everything else. PATH’s innovation scorecard specifically recommends us to collect only essential data elements as a foundational practice applicable across all settings.
This is harder than it sounds.
Vertical disease programs each demand their own data streams. Donors want their specific indicators tracked. Before you know it, a rural clinic nurse is spending three hours daily on data entry instead of seeing patients.
Creating a culture of data use requires deciding what questions you actually need answered, then ruthlessly cutting everything else. Digital tools should reduce burden, not create 47 different login credentials for parallel reporting systems.
2. Dirty Secret of Performance-Based Funding
Here’s something PATH’s expert consultations uncovered that rarely makes it into donor presentations: performance-based incentives are creating perverse outcomes. When health facilities receive funding based on reported metrics, frontline workers start “cooking” the data to hit targets rather than accurately reporting reality.
This isn’t a technology problem.
No amount of blockchain-verified, AI-validated, cloud-based data collection will fix incentive structures that reward manipulation. WHO’s health data governance principles emphasize equity and human rights, but we’re building systems that instead incentivize lying.
The PATH report found this pattern across all four countries studied.
Facilities knew exactly which numbers mattered for funding and mysteriously, those numbers improved while actual service quality stagnated. Digital tools can make this fraud more efficient, but they can’t prevent it.
3. Some Problems Just Need Money
PATH’s analysis revealed something refreshingly honest: some barriers to good data have no corresponding innovations because they don’t need innovation. They need re
- Paper shortages? Buy more paper.
- Inadequate workspace? Build better facilities.
- Lack of secure storage? Install locks and shelving.
These aren’t sexy solutions. They won’t win awards at digital health conferences. But they work.
The development sector’s obsession with “innovation” obscures this simple truth. Sometimes the most effective intervention is allocating existing re
PATH’s map-and-match tool specifically identified gaps without innovative solutions: inadequate space for data work, shortage of paper-based tools, and lack of secure storage. The honest recommendation? Just fix them with conventional re
4. The Transition Period No One Plans For
Countries across Africa are prioritizing nationwide digitalization. Ethiopia’s Information Revolution, Kenya’s eCHIS rollout, the DHIS2 implementations everywhere. These are long-term processes taking years to complete.
PATH’s critical insight: nobody plans for the transition period. Health facilities are told to use new digital systems while still maintaining parallel paper processes “just in case.” Community health workers carry smartphones for digital reporting but must also submit the same data on paper forms.
This isn’t cautious change management. It’s doubling frontline worker burden during the exact period when you need their buy-in most. PATH recommends specific interim solutions like Bluetooth-based data transfer for offline contexts and photo-to-digital tools like ODK Scan that can bridge paper and digital worlds.
Digital health implementation during COVID-19 taught us that rushed digital transformations without transition planning create chaos. Yet we keep making the same mistake, assuming frontline workers will somehow manage dual systems indefinitely.
5. Solar Panels Beat AI Algorithms
PATH’s scorecard evaluated 80+ innovations and prioritized 20 for detailed assessment. The most impactful recommendations for accelerating digitalization? Not AI, not blockchain, not predictive analytics.
Solar-powered systems for reliable electricity and internet connectivity.
For low-maturity sites especially, consistent power matters infinitely more than sophisticated software. You can have the most elegant EMR system ever designed, but if the tablet dies at noon and the facility has no electricity until tomorrow, your innovation is worthless.
This aligns with what successful digital health global goods have demonstrated: solutions need to work offline, sync peer-to-peer, and assume intermittent connectivity. Tamanu’s sync-first technology and CommCare’s offline-first design reflect this reality.
PATH specifically recommends solar-powered IoT sensors for high-maturity sites with governance infrastructure already in place. For everyone else? Get the lights on first.
6. What Banking Figured Out Decades Ago
One PATH expert consultation mentioned something striking: the financial industry solved interoperability challenges decades ago. Your Visa card works at any ATM worldwide. Banks achieved seamless data exchange and built consumer trust in data protection at massive scale.
Meanwhile, health systems still struggle with basic interoperability. WHO’s Digital Health Platform Handbook has documented the infrastructure approach, but implementation lags far behind.
The financial sector had clear incentives: transactions fail without interoperability, costing real money immediately. Health data fragmentation has diffuse costs spread across systems, making the problem easier to ignore.
This isn’t a technology gap. Banks didn’t have better programmers than health systems. They had clearer accountability for system failures and stronger economic incentives to solve them.
7. The Gender Problem Nobody Mentions
PATH’s research uncovered a pattern invisible in most digital health evaluations: gender disparities in data work reflect broader inequities in health systems. Women are concentrated in frontline clinical roles that include data collection as “extra” unpaid work. Men occupy higher-level positions with dedicated time for training and capacity building.
Digital tools don’t automatically fix this. Without explicit attention to who accesses training, who receives devices, and who gets compensated for data work, digitalization can reinforce existing gender inequalities rather than addressing them.
PATH recommends ensuring women and men in data roles have equal access to smartphones, internet connectivity, and data plans. It also suggests monitoring implementation with qualitative methods to capture how innovations differently impact women and men health workers.
This isn’t a side issue. If your digital health intervention assumes everyone has equal access to technology and training, you’re building on a foundation of inequality.
What Actually Works
PATH’s synthesis offers unglamorous but proven recommendations:
- Strengthen fundamentals first: Clear role definitions, collecting only essential data, and robust supervision matter more than sophisticated tools. These practices apply regardless of digital maturity level.
- Match solutions to infrastructure reality: High-maturity sites with reliable power can consider IoT sensors and biometric systems. Low-maturity sites should focus on solar power, offline-capable tools, and simple visual management systems like color-coded stock cards.
- Build the evidence base gradually: Many newer innovations lack effectiveness data. Countries should scale proven solutions while carefully evaluating new approaches rather than betting everything on unvalidated tools.
The PATH team created detailed implementation roadmaps, scorecards, and map-and-match visualizations precisely because effective digital health requires methodical planning, not breathless innovation.
Your health data problems are real. The solutions exist. They’re just less exciting than the pitch decks suggest.
