In 2023, we did not set out to build a great system. We asked our county software developers a small question: could our hospitals be less dependent on paper?
We were tired of mothers arriving with an exercise book, split into two to make two files. When that book got lost, her story got lost. A clinician had no history to consult. We also knew paper hides financial gaps. Money paid at the counter did not always reach accounts meant to buy medicines.
Our prayer was the patient be seen, the medicine counted, every shilling traceable.
Our developers suggested we seek permission to adapt an existing national system at Afya House. The Department of Defence and University of Nairobi had adopted it earlier for their hospitals.
We did not get its source code, only access and permission to adapt it. What we inherited was not a finished suit but cloth to tailor.
For months our team walked wards, learning how nurses handle outpatient visits, how labs queue samples, how pharmacy dispenses. A hospital is a small city. You must walk its streets before drawing its map.
Then came the uncelebrated work: training people. Thousands had to learn new ways. There was theory in a hall, then the practical at 2am in Kangema, helping a nurse admit a mother in labour into a computer for the first time.
Once live, challenges humbled us. Computers often broke. We opened a small workshop to repair what would have been discarded. The internet was humbling too. A provider promised 170 facilities online, but in the hills of Kigumo and Mathioya speeds were painfully slow. Fibre helped towns, not villages. We tried satellite connectivity, an American innovation, which worked where fibre could not. Power blinked, so we slowly started to add solar in our health facilities. Now we are at 80%.
It took nearly four years for http://Afya.ke to become habit, not project.
Then we noticed something. Over three years, our annual hospital revenue rose from about Sh80 million to about Sh800 million. At Murang'a Level 5 Hospital, daily collections rose from about Sh60,000 in cash to about Sh600,000 recorded and visible in the system.
Digitisation alone does not explain every increase, but it helped close gaps between what patients paid, what was recorded and what reached county accounts. The point was not to make patients pay more, but to ensure money already paid was accounted for. That money bought drugs.
We were not alone. Tharaka-Nithi came to us and adopted the same system in its main hospital. We were one county among others adopting Afya.ke.
As it matured, ideas came from wards. Could we link it to eCitizen so patients access records on phones? Could we put it inside MyMuranga, our county app?
Then a personal question: why should a 65-year-old grandmother with diabetes, hypertension, HIV or mental illness spend Sh2,000 monthly travelling to Murang'a Level 5 for a refill?
Telemedicine was thus born in a refurbished wing of one of our oldest hospitals. Newly recruited young doctors saw patients over video; drugs were availed closer to home. Not perfect, but it saved bus fare and hardship.
We dreamed of Tele Mama: linking pregnant mothers in remote dispensaries to specialists for earlier review. These ideas did not start in boardrooms, but from listening to nurses and clinical officers.
Kenya is pursuing a unified national digital health system, offered commercially by private developers.
We understood the need to align and be certified. As we say, mwenye nguvu mpishe.
We were told it had not met Digital Health Authority certification. Certification matters, no system should be exempt. Our concern was the disruption and failure to allow inter operability for a free public government syre.
Further, the replacement did not yet offer what we had built around local needs. Telemedicine was unavailable in same form, e-Citizen and MyMuranga links were gone, Tele Mama remained a dream. We were told capabilities would come later.
Meanwhile records could not be retrieved mid-queue, services stalled, telemedicine went dark.
I cannot forget faces that day. The head of telemedicine, a young lady who mothered the unit like her child, worried not about herself but her staff who would have no work, and grandmothers who would again board matatus they could barely afford. The management team stood like mourners, confronting loss of routines built over years.
A nurse in Kangema could once report a problem and see local developers fix it that week. With a private system, requests pass through formal processes and release schedules. Those protect quality, but can slow local innovation if they do not accommodate it.
Local ownership does not guarantee quality, commercial ownership does not guarantee responsiveness. The test is whether a system is safe, reliable, accessible and able to improve care.
http://Afya.ke was never just software. It was nurses learning new ways, youth repairing computers, patients travelling less, public money becoming visible.
We may not have built Kenya's final system, but we learned how useful systems grow: through standards, yes, but also through listening and solving problems close to where they occur.
The question remains: as Kenya builds its national system, will a nurse in a remote dispensary still shape it? Will a grandmother's needs influence its next improvement?
Governor Dr Irungu Kangata,Muranga, PhD in law .