Strengthening data systems within learning health systems in Kenya
Bibliographic record
Abstract
Introduction Many countries are working towards the Sustainable Development Goal 3.2 target of reducing neonatal mortality rate to under 12 per 1,000 live births. However, in Sub-Saharan Africa, the neonatal mortality rate remains high at 27 deaths per 1,000 live births in 2019. With more births occurring in hospitals, it is important to strengthen inpatient newborn care by improving newborn monitoring charts as a step towards improving the quality and quantity of documentation and subsequently quality of care. Documentation can be done using paper-based or electronic data systems which generate data that can be used for daily care provision as well as quality improvement. Research question: How can routine data systems be strengthened in an LMIC to support a learning health system agenda? This study was conducted within the Clinical Information Network for Newborns (CIN-N) in Kenya. The CIN-N is a network of 22 county referral hospitals in Kenya. Its aim is to improve the quality and use of information for decision making and therefore improve the patient outcomes. The network data system comprises a clerk who extracts discharge data from hospital records onto a customised database. Overall, a multi-methods approach was adopted through qualitative studies, evidence synthesis, and quantitative studies. Findings showed that the electronic medical record systems that were in place at the public hospitals, could not support the data technical block of a learning health system as envisioned for inpatient newborn care because there were no functioning inpatient modules. Therefore, a hybrid data solution that combines paper-based and electronic data system was found to be contextually appropriate. Next, the study sought to understand the global evidence on designing monitoring charts through a scoping review which showed that studies followed a general non-systematic process of designing paper-based monitoring charts. The next step of the project involved designing, piloting a newly designed monitoring chart using a Human centered design approach. The implementation process varied across hospitals, for example, some hospitals opted to train all staff together during continuous medical education sessions while others trained staff during hand-over sessions at the end of a shift. The chart was well received at the newborn wards within the network of hospitals with users citing benefits such as reduced writing, consolidated information, and improved communication. However, challenges emerged relating to the work environment and staffing, inadequate supply of charts and inadequate equipment to support monitoring tasks. These challenges also provide opportunities to improve processes within the hospital to overcome them and improve documentation. Lastly, we evaluated the documentation of key vital signs – temperature, pulse, respiratory rate, and oxygen saturation, by assessing the number of times each was documented and the number of times the set was documented in the first 48hours. This quantitative evaluation showed that all vital signs recorded an improvement with oxygen saturation recording the highest improvement. Further, sicker babies were likely to receive more frequent documentation of vital signs as is the recommended practice. However, there was still room for improvement as nearly half of the newborns did not have a single full set of documented post-admission TPRS by the end of the 48 hours study period, and there was variability in hospital performance. Conclusions While this PhD showed that the design process and technical design of the chart was important, it also illustrated that an enabling environment is crucial to ensure successful implementation and chart uptake. Hospitals, are considered as complex systems with people, processes equipment and institutions working together. Therefore, a systems perspective is required to facilitate implementation and explore emerging issues to strengthen documentation of newborn care and subsequently improve care.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.025 | 0.029 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.002 | 0.003 |
| Science and technology studies | 0.005 | 0.005 |
| Scholarly communication | 0.006 | 0.007 |
| Open science | 0.002 | 0.006 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.006 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".