CHAPTER 5: RHIS DATA MANAGEMENT AND USE

1
 Session Objectives

 Define and discuss routine health information system.

 Describe Information Cycle.

 Know data quality dimensions and data quality assurance

technique.

2
 Introduction to RHIS

 RHIS is an ongoing data collection on health

status, health interventions, and health resources.
 Routine HIS = often called Facility-based HIS =
often called HMIS.

3
 INFORMATION CYCLE
Information Cycle is a diagrammatic way of looking at information and
enables you to see the links between the different phases: collecting, pr
ocessing, analyzing, presenting, interpreting and using information.
 Data collection: Recording of health data using individual and
family folder, registers, tally and reporting formats.
 Data processing: is a process of cleaning, entering and aggregation
of data.
 Data analysis and presentation: is a process of interpretation and
comparison of generated information in the form of sentence, tables
and graphs.
 Data utilization: is the continues use of data in the activity of the
organization to improve operational efficiency.

4
 Information cycle
Tools:
Collection Client cards
 Indicator based essential dataset Registers
Use of  Data elements - std. definitions tally sheets
Information  Data sources & tools defined
 Data capture streamlined
 Regular review of data
 Relate to operational plans Tools:
Programme Management Tools:
 Monitor service coverage & Planning Processing Collation tools
quality Budgeting
Validation rules
Data flow
Tools:  Collation
Explore  Data quality checks (manual)
Interpretation Ask questions  Data validation (computer)
Research
Feedback
 Making sense of Tools:
information Tools:
 Possible interpretation Presentation Tables Analysis Indicator formulas
Graphs
 Explore Discussions

 Flow of information Feedback Indicators

 Format of tables, graphs
& reports
 Feedback mechanisms

5
 Type of data analysis Type of Chart Data Needed

Frequency of occurrence: Simple Bar chart Tallies of category (data can be attribute data
percentages or comparisons of Pie chart or variable data divided into categories).
magnitude. Pareto chart

Trends over time Line graph Measurements taken in chronological order

Run chart (attributes or variable data can be used).

Distribution: Variation not related Histograms Large number (40 or more)of measurements
to time (not necessarily in chronological order,
variable data).

Association: Looking for a Scatter diagram Large number (40 or more)of paired
correlation between two things measurements (measures of both things of
interest, variable data).

6
DATA QUALITY

7
Data quality is often defined as “fitness for use.” What does this
mean?

 Data are fit for their intended uses in operations, decision

making, and planning.

 Data reflect real value or true performance.

 Data meet reasonable standards when checked against

criteria for quality.

38
 Symptoms of Data Quality Problems

 Different people supply different answers to the same question.

 Data are not collected in a standardized way or objectively
measured.
 Staff suspect that the information is unreliable, but they have no
way of proving it.
 There are parallel data systems to collect the same indicator.

69
 Causes of Poor Data Quality
Technical determinants
 Lack of guidelines to fill out the data sources & reporting forms.
 Data collection & reporting forms aren’t standardized.
 Complex design of data collection and reporting tools.
Behavioral determinants
 Personnel not trained in the use of data sources & reporting forms.
 Misunderstanding of how to compile data, use tally sheets, and
prepare reports.
 Math errors occur during data consolidation from data sources,
affecting report preparation.
810
Organizational determinants

 Lack of a reviewing process, before report submission to next

level.

 Organization incentivizes reporting high performance.

 Absence of culture of information use.

11
 Possible Solutions of Data Quality

 Standardization and simplification of guidelines, recording and

reporting formats.
 Integration and institutionalization of health data.
 Staffing of health institutions.
 Build capacity of health work force.
 Strengthen PMT at each level of the health system.
 Enhance culture of information use.

12
DATA QUALITY DIMENSIONS

13
 1. Accuracy and Validity

 Accurate data are considered correct: the data measure what they
are intended to measure.

 Accurate data minimize error (e.g., recording or interviewer bias,

transcription error, sampling error) to a point of being negligible.

 The original data must be accurate in order to be useful.

14
 2. Reliability (Consistency)

 Data should yield the same results on repeated collection,

processing, storing & display of information. In other words, data
should be consistent.

15
 3. Completeness
All required data should be present & the medical record should contain all
pertinent documents with complete and appropriate documentation.

Data Completeness on data recoding tools

 All necessary data elements on registers/forms/cards should be filled

immediately after provision of the service by the care provider.

 The cover page of folder should contain all the necessary identifying data to
uniquely identify an individual client.

 For inpatients or clients received the service, the registers should contain all
necessary information’s accurately pertinent to the service provided and those
include on registers.

 For all medical records, relevant forms are complete, with signatures and date of
attendance.
16
 This refers the extent to which facility and woreda filled all data
elements in the reports or data base for all reportable events.

 Health facilities are expected to fill a zero value in the reporting

form even if the event doesn’t happen in a defined reporting
period.

17
 4. Timeliness

 Information, especially clinical information, should be

documented as an event occurs, treatment is performed or
results noted.

 Delaying documentation could cause information to be omitted

and errors recorded.

18
 5. Legibility

 All data whether written, transcribed and/or printed should be

readable.

 In all medical records, cryptic codes or symbols cannot be used in

either manual or electronic patient records.

 If abbreviations are used, they are standard and understood by all

health care professionals involved in the service being provided to
the patient.

 Mostly this problem is seen at outpatient and inpatient department

which are major source for clinical data and NCoD.

19
 6. Accessibility

 All necessary data are available when needed for patient care &
for all other official purposes.

20
 7. Precision

 means the data have sufficient detail. E.g. an indicator requires

the no. of individuals who received HIV counseling & testing &
received their test results by sex of the individual.

 An information system lacks precision if it is not designed to

record the sex of the individual who received counseling and
testing.

21
 8. Integrity

 Integrity is the quality of being honest and having strong moral

principles or moral uprightness.
 Data Integrity can be considered as a polar opposite to data
corruption that renders the information as ineffective in fulfilling
desired data requirements.
 Data integrity aims to prevent unintentional changes to
information.
 Data have integrity when the systems used to generate them are
protected from deliberate bias or manipulation for political or
personal reasons.
22
 9. Confidentiality

 clients are assured that their data will be maintained according to

national and/or international standards for data.
 Or personal data are not disclosed inappropriately, & that data in
hard copy & electronic form are treated with appropriate levels of
security.
10. Relevance
The level of consistency between the content of data and user’s area
of interest.

23
WHAT IS DATA QUALITY ASSURANCE?

24
 A systematic monitoring and evaluation of data to uncover
inconsistencies in the data and data management system,
and making necessary corrections to ensure quality of data.

25
DATA QUALITY ASSURANCE METHODS

26
 Supportive supervision/self-assessments

 Independent assessments/audits

27
 Independent assessments/auditing

Driven externally, often by funding agency.

Standard and structured.
Conducted by external audit team.
Capacity-building is not primary objective.

28
 Supportive supervision/self-assessment

Driven internally.

Flexible.

Conducted by program or district managers using its M&E system.

Encourages dialogue to set data-quality improvement priorities.

Objective is to develop actions that will strengthen program.

29
DATA QUALITY ASSURANCE TOOLS

30
1. LQAS (Lot Quality Assurance Sampling )
 Self-assessment at facility level.
 Simple and uses small sample size for continues quality
assurance at facility level.
 can be used through data accuracy check lists.
 limited to few data quality components (mostly accuracy).
 Data elements selected randomly.
 Frequency: Monthly.
 Data quality dimension addressed.
 consistency of reported data.

31
 It is a technique useful for assessing whether the desired level
of Data accuracy has been achieved by comparing data in relevant
record forms (i.e. registers or tallies) and HMIS reports.
 will be used to check data accuracy at Health Facility level.
 Health Facilities will maintain a registry to record the data
accuracy and check results and to look the trend of the data
quality improvement.

32
 Steps to carryout LQAS

1. Decide the month for which you want to do the data accuracy check (HF are
expected to do Monthly).

2. Pre-fix the level of data accuracy that you are expecting, e.g. 70% or 85% etc.

3. Put serial numbers against the data elements (sum section) not disaggregation
in the Service Delivery or Disease Report that you want to include in the data
accuracy check (E.g.: New acceptor, Repeat acceptors not age or method
disaggregation).

4. Randomly select 12 data elements.

5. List down the selected data elements from the report on to the data accuracy
check sheet in Column 2 & 3.

33
6. Write down the reported figures from the Monthly HMIS Report for the
selected data elements in the Column 4 of the Data Accuracy Check Sheet.

7. Recount the figure from the corresponding registers and note the figures on
Column 5 of the LQAS check-sheet.

8. If the figures for a particular data element match or do not match put “yes” or
“no” accordingly in Column 6 or Column 7 respectively.

9. Count the total number of “yes” and “no” at the end of the table.

[Link] the total number of “yes” with the LQAS Decision Rule table and
determine the level of data accuracy achieving the expected target or not.

34
Data Accuracy Check Sheet
Month/Quarter/Year: --------------

35
LQAS DECISION RULES

36
 2. Routine Data Quality Assessment (RDQA)

 RDQA is an assessment tool that can be used to self-monitor

assessment and progress and evaluate the RHIS status.
 units to verify reported data against to source documents and to
look RHIS system implementation.
 It is recommended to be implemented quarterly by administrative
health unit.

37
 Sampling Methodology

For Regional level:

In regions with zones:

 Randomly select 4 zones.

 From each of the selected zones, randomly select three woredas.

 From selected woredas, select randomly one HC/Hosp.

38
In Regions without zones

 Randomly select 4 woredas.

 From selected woredas, randomly select three HC/Hosp.

b. For zonal level

 Randomly select 4 woredas.

 From selected woredas, select randomly three health centers or

hospitals.

c. For woreda level

 Use census of all health centers and hospitals in the woreda.

39
RDQA STEPS
1. To understand the HMIS data accuracy level in the given administrative
area, a sample of 12 health facilities is selected.
2. Each of these selected health facilities is visited to complete the RDQA
tool.
3. Select key data elements from the HMIS reports that will be studied
(include data elements of 7 to 9 top priority indicators at national
level).
4. List the data items in the RDQA table.
5. For each of the selected data elements recount the number of cases or
events recorded during the reporting period by reviewing the relevant
source documents available at the selected sites[A].

40
6. Copy the number of cases or events for the selected data elements
reported by the site during the reporting period from the HMIS
reports submitted by the selected sites [B].

7. Add up all the recounted figures for the corresponding data

elements from the 12 sites [∑A].

8. Add up all the figures for the same data elements copied from the
HMIS reports of all the 12 sites [∑B].

9. Calculate the ratio of recounted to reported numbers. [∑A / ∑B].

41
42
 According to WHO RDQA the acceptable range of DV is
between 90% - 110%.
DV < =90% is over reporting.
DV >=110% is under reporting.
From the Example:
 ANC 4th Visit, Pentavalent 3 & Meseals = over-reporting.
 Currently on ART & TB all forms = under-reporting.
 Skilled delivery = acceptable report.

43
FREQUENCY

Level Data verification Full RDQA

FMOH Bi-annually Annually

RHB Quarterly Annually

WoHO Quarterly Annually

Health Facilities NA NA

44
 3. DQA

 Assessment by funding agency.

 Standard approach to implementation.

 Conducted by external audit team.

45
 4. PRISM

 To assess whether technical, behavioral and organizational

determinants have influence on RHIS performance.

46
 5. Data Quality Desk review

 The desk review of data quality is a review of the quality of

existing data from RHIS hence, it does not require additional data
collection.

 Frequency: As many of the consistency metrics require annual

data, the FMOH also recommends conducting this review
annually.

 Data quality dimensions addressed.

 consistency of reported data

 Completeness

 Timeliness. 47
 6. Visual Scanning (Eye Balling)

 It is a simple method used at HF to check for consistency of

reports before/after conducting data entry.

 The PMT members sit together and look across each line and
then from top to bottom to identify:

 Missing data values.

 Unexpected fluctuations beyond max/min values,

 Inconsistencies b/n linked data elements.

 Mathematical errors.

48
 Frequency: Whenever report is generated.

 Data quality dimensions addressed:

Presence of outlier.

Data completeness.

consistency b/n indicators.

49
50