Assessment of data quality and information use of the community health information system: A case study of Karurumo community health unit- Embu county, Kenya
View/ Open
Date
2015Author
Ndegwa, Caroline W
Type
ThesisLanguage
enMetadata
Show full item recordAbstract
This study focused on assessing data quality and information use of Karurumo Community
Health Unit (CHU) data collected through the Community Health Information System
(CHIS), for the period September to November 2014. The assessment focused on three
dimensions of data quality: completeness accuracy; and timeliness. Use of information was
assessed based on: dialogue, action, feedback, referral, sharing and advocacy. Data was
obtained from the CHIS data collection tools as well as from the District Health Information
Software (DHIS2). Completeness as the degree to which the Community Health Information System’s data covers all services and is filled out in full on data collection forms was assessed at three
levels: Completeness of Community Health Volunteers reporting, assessed the total number
of reports received from Community Health Volunteers in relation to the total number of
reports expected from the Community Health Volunteers. Data element completeness
assessed the total number of reports from the Community Health Volunteers with no
missing values. While Completeness of Community Health Extension Workers reporting
assessed the total number of data elements filled by the Community Health Extension
Worker for the month in relation to the total number of data elements expected to be filled
for that month. The results of the assessment show that Completeness of reporting for the three months being assessed was excellent, with an average score of 95.4 percent. Community Health
Volunteers reporting was very good, with a score of 90 percent. The Community Health
Volunteers data element completeness was excellent with a score of 100 percent. The
completeness of Community Health Extension Workers reporting was also excellent with a
score of 96.4 percent. Accuracy as the matching of data transmitted from one level to another in the Community Health Information System was assessed at three levels by selecting four data elements to
be assessed. At the Community Health Volunteers reporting, the data elements were
aggregated and compared with the Community Health Extension Workers summary. The
next level of assessment was to compare the data elements in the Community Health.
Extension Workers summary against the same data elements in DHIS2. The third level of
assessment for accuracy was to compare the Community Health Extension Workers
summary report with the chalkboard report. The results show that the accuracy of aggregated data from the Community Health Volunteers service delivery log book compared to the way it was reported in the
Community Health Extension Workers summary reports was excellent, at 98.8 percent,
while that of the Community Health Extension Workers summary report compared to the
DHIS2 report was not calculated, since none of the Community Health Extension Workers
summary reports for the three months were entered into DHIS2. The same applied to data
transmission from the Community Health Extension Workers summary to the chalk board,
since the chalk board was last updated in June 2014 Timeliness can be defined as the extent to which data are sufficiently up-to-date for a task. Information is timely if information is available when needed. For this study, keeping deadlines for the submission of reports was used as a proxy measure of the timeliness of the data. This was to be measured by checking the date of receipt of reports from
Community Health Volunteers by Community Health Extension Workers for the three
months and recording those that were received before and after the deadline. From the
study, it was realized that the Community Health Extension Workers does not record the
date of receipt of reports from the Community Health Volunteers, since the reporting form
does not have such a provision. Therefore, it could not be determined whether reports
were received before or after the submission deadline. This means that timeliness could
not be established for the Community Health Unit. Use of information was assessed by allocating two points to every answer that was YES and zero points to every NO answer. There were thirteen questions, and the highest possible score was twenty six points. Scores between 2 and 8 indicate that use of information overall is very weak. Scores between 10 and 14 indicate weakness but some signs of use of
information. Scores 16 through 20 indicate beginning of active use of information, and 22 to 26 indicate an advanced level of use of information. For the study, Use of Information scored 20 out of 26 points (76.9 percent), indicating the beginning of active use of information. Dialogue scored 2 points out of a total score of 4. Action scored 2 out of 4 points; feedback scored 4 out of 6 points; referrals scored 6 out of 6; sharing scored 4 out of 4; while advocacy scored 2 out of 2. The recommendations of the study for Policy and Programmes are as follows: policy
makers should come up with a system to ensure all the Community Health Units conduct
routine data quality assessments, and develop action plans to improve the quality of data to
inform decision making. A second recommendation is that data collection tools need to be reviewed to ensure that they capture the timeliness of the data collection and transmission processes. This can be
done by including a place for indicating the date of receipt of reports at the higher levels.
The transmission of data from the manual CHIS forms (MOH 515) into the electronic form
in DHIS should be a practice that is inculcated in all the CHUs, since it is a requirement for
reporting. Programmes that are using the CHIS can also conduct assessments of the CHIS to establish
a baseline of CHIS performance for improvement. The findings will serve as a basis of
comparison with information collected in later assessments. Assessment of the CHIS will
thus be a quality improvement strategy for the CHIS. Financial and technical support for dialogue and action days should be provided, to ensure that all the Community Health Units are functioning as required and also accomplish their set mandates. For the referral system, the clinicians in the health facilities that attend to the clients that have been referred from the Community Health Units need to be sensitized on the need to complete the referral forms, so that the referrals are completed.
Recommendations of the study for future research are as follows: there is need for further research to identify the factors that influence the quality of data of the Community Health Information System, to enable the users of the system to identify the areas that need improvement and to come up with action plans for data quality improvement. Future studies can also focus on other dimensions of data quality including: validity, periodicity, relevance, reliability, precision, integrity, confidentiality, comparability, consistency, concordance, granularity, repeatability, usability, objectivity, accessibility, transparency and representativeness. Since Karurumo is considered a model CHU, the data quality and information use practice observed in the CHU are not the norm. future studies can therefore compare a model CHU and one that is not enjoying donor support to establish whether the data quality and information use practice of such a CHU are similar to those of a model CHU.
Publisher
University of Nairobi