Data Quality Audit Report

8TH ROUND

Based on the data quality audit visits in 7 districts of Uttar

Pradesh during 29 March-1 April 2022

PREPARED BY
UTTAR PRADESH NATIONAL HEALTH MISSION

SUPPORTED BY
UTTAR PRADESH TECHNICAL SUPPORT UNIT (UP TSU)
1
Major Contributors

1. Dr Anamika Misra
General Manager- Monitoring and Evaluation, NHM

2. Dr Ravi Prakash
Sr. Deputy Director- Monitoring and Evaluation, UPTSU

3. SVP Pankaj
Deputy General Manager- Monitoring and Evaluation, NHM

4. Anand B. Tripathi
Senior Specialist- Monitoring and Evaluation, UPTSU

5. Huzaifa Bilal
Senior Specialist- Monitoring and Evaluation, UPTSU

6. Dr Shivanand Chauhan
Specialist- Monitoring and Evaluation, UPTSU

2
LIST OF ABBREVIATIONS

AdRO Additional Research Officer

ANC Ante Natal Care
CH Child Health
CHC Community Health Centre
CMS Chief Medical Superintendent
CMO Chief Medical Officer
DH District Hospital
DEO Data Entry Operator
DCH District Combined Hospital
DPM District Program Manager
DWH District Women Hospital
DG FW Director General Family Welfare
DG MH Director General Medical Health
FP Family Planning
FRU First Referral Unit
HEO Health Education Officer
HM Hospital manager
HMIS Health Management Information System
M&E Monitoring and Evaluation
MH Maternal Health
MO I/c Medical Officer In-charge
RSK Rogi Sahayta Kendra
SHI State Health Index
SN Staff Nurse
JSSK Janani Shishu Suraksha Karyakakram
UPHMIS Uttar Pradesh Health Management Information System
UPNHM Uttar Pradesh National Health Mission
UPTSU Uttar Pradesh Technical Support Unit

3
Content

1. Executive summary……………………………………………………………………………5
2. Background……………………………………………………………………………………7
3. Objectives of the data audit ………………………………………………………………. ….8
4. Methodology
4.1 Audit area and audit team …………………………………………………………9
4.2 Process of data audit…………………………………….………………………....11
4.3 Tool used for data audit ……………………………………………………….......11
4.4 Data & period of audit…………………………………………………………......12
5. Data audit findings…………………………………………………………………………….12
6. Major challenge…………...…………………………………………………………………....19
7. Suggestive solutions……………………………………………………………………………20
8. Glimpses of data audit…………………………………………………………………….……22

4
1. EXECUTIVE SUMMARY

Data quality audit is a supportive supervision approach to identify the data quality gap and suggest
corrective action for data quality improvement. Given the same, the state data quality audit team was
constituted by MD NHM in January 2018 in compliance of government order issued by the Principal
Secretary in May 2017 (संख्या- 35/2017/303/पां च-9-2017-9(127)/12). Eight rounds of audits were
conducted by the team in 134 district and block-level facilities (49 DWH/DCH & 85 CHCs) of 50
districts till March-April 2022 to understand the improvements in HMIS data quality and persisting
gaps.

The recent audit was conducted in 21 facilities (15 block facilities and 6 DH facilities) across seven
districts which are Amethi, Ballia, Bareilly, Hardoi, Jalaun, Kaushambi & Sambhal from 29th March-
1st April 2022. State data audit is a quarterly activity but it could not be planned after Feb 2020 due to
the Covid-19 in the state. The data audit was conducted with the help of the revised structured tool,
which was done during the 6th round of data audit, comprised of 66 critical data elements covering
antenatal care, delivery/newborn care & complication, family planning, child health, mortality details
and hospital services. This covers all the data elements of ranking and NITI Aayog’s SHI indicators
with a few additional critical indicators of state priority. In total seven teams were formed for the audit
which comprised of members from DGMH, DGFW, NHM and UPTSU.
In total, 24 common data elements (from DH and CHC) were common across all the rounds of audit.
There has been a continuous increase in matching of portal data with source documents1 from first to
seventh round. The overall matching of 24 data elements for borth CHC & DH increased from 57%
in the first round (Jan 2018) to 74% in the 7th round (Feb 2020) of data audit. However, a slight
decrease in data matching was observed during the 8th round which took place two years later than
the 7th round. The overall matching of 24 common data elements with source document was found
to be 69% in this round (8th round).
While there has been a general improvement in data quality, a few facilities and programdomains
continued to show data quality issues. Some of the most common reasons for data quality issues
included poor and non-uniform availability of source documents (only 54% of data elements of four
major domains were having a provision in registers to record the information), printing of source
formats( More than one third of CHC facilities still do not have printed HMIS and UPHMIS format),
and no designated nodal to review data availability and quality ( 14% CHC facilities did not assigned
nodal to review the data and its quality).
Non-functionality of the validation committee is one of the major bottlenecks observed during the
data audit. It was observed that more than one third (35%) of the visited facilities (block facility and
DH) did not conduct validation meetings during the last quarter. Lack of understanding of some of

1
Data elements reported value matched with the value recorded in the source document. Deviance within 10 percent in
the value from source document has been considered as matched for all data elements except mortality (infant, child and
maternal death).

5
the data elements (maternal and newborn complications) was also identified as one of the reasons for
low data quality during the supportive supervision process with facility staff.
Training of block officials (ARO/HEO/BPM/HM/DEO), staff nurses, and ANMs emerged as one
of the main factors affecting data quality. There are only two third facilities (65%) where at least two
staff have received training and only one third (33%) ofSNs and ANMs have received training on
HMIS/UPHMIS format definition and compilation during the last one year.

Based on the gaps identified, the action plan was developed for each of the audited facilities and shared
with the facility in-charge for corrective actions. The action plan includes the gaps, suggestive
actionable points, the person responsible, and the timeline. The feedback meeting was also held with
all the blocks and findings were shared for overall improvement in the data quality of a district.

6
2. BACKGROUND

The availability of good quality data is critical for any program reviews, planning and prioritization.
Uttar Pradesh has developed and implemented a robust data system which provides a holistic platform
to obtain all the critical data required for the identification of low performing indicators, low
performing geographies and factors associated with low/high performance of indicators.

In this regard, monthly facility wise government data portals (HMIS/UPHMIS) are the primarily
reliable source for data use at all levels of the health system. Thus, ensuring availability of high-quality
data is the key. Additionally, the UP Health dashboard (district and block ranking based) has also been
developed based on HMIS/UPHMIS data and used by the health officials at different levels for review
and planning of health programs. Recognizing the criticality of reporting quality data, the state has
initiated the concept of data quality audit to improve the quality of data availability under the
government data system (HMIS/UPHMIS).

Data quality audit is a supportive supervision approach to improve the data quality of the government
data system by assessment of data quality issues at the facility level and suggesting corrective actions.
This process includes gap identification, joint problem-solving, handhold support and capacity
building. The primarily includes validating the reported data with the source document, identifying the
gaps and developing the capacity of facility staff on reporting accurate data.

The state data quality audit team was constituted in January 2018 and eight rounds of audit have been
conducted by the team in 150 facilities in50 districts till April 2022.

Data quality framework of factors affecting data quality

The complete process of correct reporting of data from service delivery to the portal can be classified
into 3 steps process, a) Data recording, b) Data transfer and c) Data entry. There are multiple factors at each
step which may affect the process to ensure the reporting of correct data. The gap in any of the
components at any step may affect the reporting of quality data (Figure 1, Data Quality Framework)

7
Figure 1 Data Quality Framework

It is, therefore, important to understand the issues and challenges at each step so that effective
measures could be taken to strengthen the data quality.

3. OBJECTIVES OF DATA AUDIT

The overall goal of the data audit activity was to ensure the availability of quality data for decision
making. Keeping in view the issues and challenges of data quality in HMIS/UPHMIS in the state
following objectives have been decided for the audit activity:
1. To validate and improve the data quality of key critical data elements
2. To assess the system level gap in the reporting of quality data
3. To assess recording and source document availability for key critical data elements

8
4. METHODOLOGY

The state had issued a letter (पत्ां क- SPMU/NHM/M&E/2021-21/25/8339-2) dated March 25,
2022, for data audit visits by the state team in the month of March-April 2022.
4.1 Audit area and audit team
Seven teams were constituted for audit in seven new districts comprising members from NHM,
Directorate and UPTSU. The list of districts and details of the team are given below in Table 1.

Table 1: Data Quality Audit Team

Team Team Members Department Date of Selected

Visit District
Team 1 Mr. Ved Prakash , ADRO DGFW 29 March
th Sambhal
Mr. Prashant Srivastava, Divisional NHM to 01st
Consultant–QA April, 2022
Mr. Nazir Haider, M&E Specialist UPTSU
Team 2 Mr. Yogesh Chandra,ADRO DGFW 29th March Amethi
Mr. Indrajeet Singh, Consultant, RKSK SPMU, NHM to 01st
Ms Yogita Kharkwal, M&E Specialist UPTSU April, 2022
Team 3 Mr. Virendra Pratap (ARO) RHFWTC, LKO 29th March Ballia
Mr. Sonu Gautam, PC (M&E) SPMU NHM to 01st
Dr. Prahlad, M&E Specialist UPTSU April, 2022
Team 4 Dr. Arvind Kumar, JD (Paramedical) DGMH 29th March Jalaun
Mr. Manoz Sharma, ADRO CMO office LKO to 01st
Moh. Ajam Khan, M&D Officer SPMU, NHM April, 2022
Mr. Ved Prakash, M&E Specialist UPTSU
Team 5 Mr. Dinesh Kumar, ADRO CMO office LKO 29th March Bareilly
Dr. Raees Ahmad, Tech Consultant MH SPMU, NHM to 01st
Mr. Ankit, M&E Specialist UPTSU April, 2022
Team 6 Dr. Y.K. Pathak, JD(Blindness) DGMH 29th March Hardoi
Mr. I.C. Verma, ADRO CMO office LKO to 01st
Mr. Manish Soni, Consultant FP SPMU NHM April, 2022
Mr. Sourabh Roopchandani, , M&E UPTSU
Specialist
Team 7 Dr. Vikas Singhal, JD (Communicable DGMH 29th March Kaushambi
disease) to 01st
Mr. Uma Shankar Shukla, ADRO DGFW April, 2022
Dr. Raj Kumar Verma, Consultant, RI SPMU, NHM
Mr. Om Prakash, M&E Specialist UPTSU

9
Seven new districts namely Amethi, Ballia, Bareilly, Hardoi, Jalaun, Kaushambi & Sambhal were
constituted for the eighth round of data audit. The districts were selected based on the following
criteria:

 One district – Random selection among top 5 in district performance ranking (Feb 2022)
 One district- Random selection among the bottom 5 in district performance ranking (Feb 2022)
 Five districts – Randomly selected

Further, two block facilities and one district hospital were chosen for the audit in each district. The
block facilities were identified based on the reporting of non-zero data elements. One good
performing and one poor-performing block facility were selected for the audit. District Women
Hospital (DWH) or District Combined Hospital (DCH) as per availability in the district was selected.
As there is no district hospital in Amethi, a higher level facility at headquarter was included for data
audit. This exercise has been done by the state and the list of facilities was shared with the data audit
team.

Thus, in total, 21 facilities (15 block facilities and 6 DH facilities) from 7 districts were identified and
audited during the process.

Table 2: List of district hospital facilities selected for audit

Sr. District Block Facility Facility Type of
No. HMIS code facility
District Female Hospital
1 Ballia Ballia Sadar 407264 DWH
Ballia
2 Bareilly Bareily DHQ District Female Hospital 429512 DWH
3 Jalaun Jalaun DHQ DH District Women Hospital 354432 DWH
District Women Hospital
4 Hardoi Hardoi DHQ 384295 DWH
Hardoi
DH District Combined
5 Sambhal Sambhal DHQ 457004 DCH
Hospital
6 Kaushambi Kaushambi DHQ District Combined Hospital 400034 DCH

Table 3: List of block facilities selected for audit.

Sr. District Block Facility Facility HMIS Type of
No. code facility
1 Amethi Musafirkhana CHC Musafirkhana 412106 CHC
2 Amethi Gauriganj CHC Gauriganj 412108 CHC
3 Amethi Jagdishpur CHC Jagdishpur 412107 CHC
4 Ballia Bellhri CHC Sonwani 407243 CHC
5 Ballia Garwar CHC Ratsar 412249 CHC

10
 Sr. District Block Facility Facility HMIS Type of
No. code facility
6 Mudia Nawi Bux
Bareilly CHC Mudia Nawi Bux 400931 CHC
(Riccha/Damkhauda)
7 Bareilly Faridpur CHC Fareedpur 400933 CHC
8 Jalaun Kuthound CHC Kuthound 329826 CHC
9 Jalaun Madhougarh CHC Madhougarh 329853 CHC
10 Hardoi Ahirori CHC Ahirori 384270 CHC
11 Hardoi Madhouganj CHC Madhoganj 384288 CHC
12 Sambhal Asmoli CHC Manauta 462398 CHC
13 Sambhal Bahjoi CHC Bahjoi 425579 CHC
14 Kaushambi Kara CHC Kara 458063 CHC
15 Kaushambi Sirthu CHC Sirathu 400030 CHC

4.2 Process
The data quality audit is a supportive supervision approach to improve the data quality of the
government data system (HMIS/ UPHMIS). This process includes hand hold support, joint problem-
solving and capacity building.
The major steps to conducting the data quality audit include the following:

 Identification of facilities to be audited

 Visit and conduct audit: The audit includes the matching of the reported data value in
HMIS and UPHMIS with source documents and identifying the reasons for identified
gaps if any.
 Preparation and sharing of an action plan based on data quality issues identified with the
facility in charge. The action plan for each of the audited facilities was developed and attached
as annexure 2.
 Feedback meeting with all the concerns responsible for reporting.

4.3 Tool used for data audit

A structured tool comprised of 61 critical data elements was used for the eighth round of audit. It
covers the following domains (Table 4):
Table 4 Domains covered in data quality audit checklist

# Domain # of data elements from

HMIS and UPHMIS
1 Antenatal care 10
2 Delivery/newborn care & complication 21
3 Family planning 4
4 Child health 7

11
 # Domain # of data elements from
HMIS and UPHMIS
5 Mortality details 6
6 Hospital services 13
Total 61

The data elements were selected considering indicators recommended by NITI AYOG’s state health
index, district/ block ranking, and current program priority.
The original tool captured information on domains like human resource, training, drugs & supply and
on critical data elements. In total, there were 115 data elements. This tool was revised during 6th round
of data audit after incorportating the finding of previous rounds of audit. The number of data elements
in revised tool reduced to 61 and the revised tool focused more on critical data elements along with
system level information and source document availability. This primarily includes format availability,
validation committee, summary preparation, person responsible etc. A separate section was added to
source document availability to understand the variation and availability of records across different
facilities. The tool is attached as Annexure 1
The data quality assessment of data collected on the tool was done on four major parameters defined
below:
- % of matched- Data elements reported matched with the value recorded in the source document.
Deviance within 10 percent in the value from source document has been considered as matched
for all data elements except mortality (infant, child and maternal mortality).
- % of over reported- Reported value of the data element is greater than the value recorded in the
source document.
- % of under-reported- Reported value of the data element is less than the value recorded in the
source document.
- % of not able to audit- Data elements for which the team was not able to audit source documents
were not available at the facility

4.4 Data and period of audit

HMIS and UPHMIS reported data on HMIS & UPHMIS portals respectively for February 2022 were
decided to be audited.

5. DATA AUDIT FINDINGS

A. Comparison over different rounds

There are 24 data elements which were common across all eight rounds and were compared to
understand the change in data quality status across different rounds of data quality audit. These data

12
elements are spread across 7 different domains. The summary of the data audit over different rounds
is given below in Fig 2:

Fig 2 Trend of matching of reported value with source document

% of reported data matched with source document over

different rounds
80 74
69
62 65 65 64
70 60
57
60
50 Same set of facilities (24 in 8 aspirational districts)
40 were selected in round 3 and round 7
30
20
10
0
Round 1 Round 2 Round 3 Round 4 Round 5 Round 6 Round 7 Round 8
(Jan 18) (Apr 18) (Jul 18) (Oct 18) (Feb 19) (Aug 19) (Feb 20) (Mar-Apr
22)

There has been a continuous increase in matching with source documents from round first to seventh.
It increased from 57% in the first round to 74% in the 7th round of data audit. However, a slight fall
in matching was observed during the 8th round that was done after a gap of two years from the previous
round.
B. Comparison over sixth, seventh & eighth rounds
Common data elements (49) across all CHC and DH facilities were examined. The overall matching
with the source document decreased slightly from 67% in the 7th round to 64% in the 8th round but
was still better than matching from the 6th round (58%). The matching was higher for data elements
on delivery & outcome, mortality and newborn health. However, the matching of the reported data
with the source document was low for ANC and maternal & newborn complications. The matching
of data elements with source documents improved across all domains except ANC, maternal
complication & delivery & its outcome.

13
Figure 3.1: Trend of matching with source records over different domains

% OF REPORTED DATA MATCHED WITH SOURCE DOCUMENT FOR ALL FACILITIES

Round-6 Round-7 Round-8

96 98 93 99
92
79 83 85 79
71 67 64
64 65 62 65 62 66
54 50 51 54 55 56 61 58
48 46 46
41

Ante natal care New born Maternal Hospital Chiild health Family planning New born Delivery and its Mortality (7) Over all (54)
(10) complication complication services (2) (9) (5) health (4) outcome (4)
(5) (8)

Figure 3.2: Trend of matching with source records over different domains for DH facilities

% OF REPORTED DATA MATCHED WITH SOURCE DOCUMENT FOR DH FACILITIES

Round-6 Round-7 Round-8
100 97
92 93 90 97
86 86 88 89
71 71 76 71 80 80
74
70
58 63 63 68
55 57 53 55
51 53 51

29

Ante natal care Chiild health Family planning Hospital New born Maternal New born Delivery and its Mortality (7) Over all (54)
(10) (9) (5) services (2) complication complication health (4) outcome (4)
(5) (8)

Figure 3.3: Trend of matching with source records over different domains for DH facilities

% OF REPORTED DATA MATCHED WITH SOURCE DOCUMENT - FOR CHC FACILITIES

Round-6 Round-7 Round-8

98 100
93 93 93
81 83
72 71 76 73
62 62 62 64 66
60 56 59
54 54
44 47 49 46 45 48
39 40 41

Maternal New born Ante natal care Hospital Chiild health Family planning New born Delivery and its Mortality (7) Over all (54)
complication complication (10) services (2) (9) (5) health (4) outcome (4)
(8) (5)

14
However, considerable improvements in matching data with source documents was observed in DH
facilities over the last three rounds; it improved from 68% in the 7th round to 74% in the 8th round
(Fig 3.1). All domains at DH have shown improvement except in the delivery & its outcome domain.
The improvement in data matching across majority of the indicators in DH as compared to CHC may
be attributed to more supportive supervision visits by Divisional M&E Officers (NHM) and District
M&E Specialist (TSU) at DHs (50%) as compered to CHCs (29%) by the divisions (Indicator 7 in
Table 6).
Besides matching, many of the data elements were also found to be over-reported and under-reported.
Antenatal care and family planning are domains where some over-reporting erewere observed.
Interestingly, the newborn complications, maternal complications and hospital services were the
domains where significantly under-reporting was observed. Also, some of the data elements from
antenatal care and child health domains were not even able to be audited due to the non-availability
of documents at facilities. The details are given in Table 5.
Table 5 Data audit summary
Domain % of matched with % of over reported % of under- % of not able to
source reported audit
Roun Round Roun Roun Roun Round Round Roun Roun Roun Roun Roun
d-6 -7 d-8 d-6 d-7 -8 -6 d-7 d-8 d-6 d-7 d-8
Ante Natal
48 54 50 15 30 21 16 12 11 20 5 19
Care (10)
Delivery and
96 98 93 0 1 2 4 1 5 0 0 0
Outcome (4)
Maternal
Complication 46 64 55 23 12 13 27 24 32 3 0 0
(7)
Newborn
79 83 85 11 7 12 5 10 2 5 0 2
Health (3)
Newborn
Complication 41 51 54 21 8 14 30 40 32 8 1 0
(5)
Child Health
46 62 65 4 17 6 5 2 10 46 18 19
(7)
Family
62 66 71 20 18 18 8 14 6 10 3 5
Planning (4)
Hospital
65 56 61 6 19 7 21 21 26 8 4 6
Services (3)
Mortality (6) 79 92 99 1 4 0 5 3 1 15 1 0
Over All (49) 58 67 64 12 15 10 14 14 13 15 4 12

15
 C. Assessment of process related gaps
There are many factors that affect the reporting of quality data (Figure 1, Data quality framework). It is
essential to have these components in place at a facility for reporting quality data. The revised checklist
also captured the different factors (availability of correct format, validation committee meeting, nodal person for
data reporting, training etc) which can affect the data quality of the facility. System-level gaps over the last
three rounds are given in Table 6.
Table 6: Percentage of facility reported system-level gap over six, seven & eight rounds of data
audit
Round-6 Round-7 Round-8
S.no. Particulars
(N=21) (N=24) (N=20)
1 % of facility with appropriate printed HMIS format 67 88 75
2 % of facility with appropriate printed UPHMIS format 62 75 70
% of facilities where a nodal is assigned to review the data
3 90 83 90
and its quality
4 Training on HMIS/UPHMIS during last one year
% of facilities where at least two staff among
4.1 ARO/HEO/BPM/HM/DEO have received training 29 58 65
on HMIS/UPHMIS format definition and compilation
% of SN/ANM trained on HMIS/UPHMIS format
4.2 37 38 33
definition
5 Last validation meeting conducted
% of facilities where VCM held in last quarter
5.1 62 79 65
including current month
5.2 % of facilities where VCM never held 14 4 0
6 Use of HMIS & UPHMIS
% of facilities where facility staff was comfortable using
two or more modules of HMIS & UPHMIS (HMIS
6.1 standard & live report download/data quality app/Excel 81 88 80
import/Report download-UPHMIS customized report/Pivot table)
(Yes/No)
% of facilities where any data quality supportive
7 supervision visit/audit done in the past six months by 43 21 35
district, division or state officials (Yes/No)
7.1 DH 43 25 50
7.2 CHC 43 19 29

C.1 Format availability

Availability of correct format is essential at the facility to collect the required information. A majority
of the facilities have appropriate printed HMIS and UPHMIS formats. A slight fall in the availability
of format has been observed in the current round compared to the previous round. One-fourth of the

16
facilities still do not have printed HMIS formats and more than one fourth (30%) percent haven’t
printed UPHMIS format.
C.2 Nodal to review the data and its quality
Considerable improvement has been observed for the assignment of a nodal to review the data & its
quality. But there are still 10% of facilities that have not assigned a nodal. BPM/ARO at block facility
and Hospital manager at district hospital are working as data nodal at a majority of the facilities.
C.3 Validation committee meeting
The validation committee was constituted to validate the reported data and ensure the quality of
reported data. The validation committee meeting is to be held monthly. However, it was observed that
one-third of the facilities haven’t conducted the meeting during last quarter.
C.4 Training on HMIS/UPHMIS
There has been considerable improvement in the training of block officials but concerns for SNs &
ANMs persist. Two third of block officials and only one-third of SNs and ANMs have received
training on HMIS/UPHMIS format definition and compilation during the last one year.
C.5 Skill to use portal
The ability to use different modules especially downloading HMIS standard & live reports from the HMIS
portal, using data quality application, Excel import, Report download-UPHMIS customized report and Pivot table on
UPHMIS portal were observed by the demonstration by Hospital manager/BPM and DEO. A slight fall has
been observed in the skill compared to the previous round. Hospital managers/BPM and DEO are
comfortable using two or more modules of HMIS & UPHMIS in 80 percent of facilities. In the case
of three or more modules, the Hospital manager/BPM and DEO are comfortable to use only 60% of
the facilities during the eighth round of data audit.
C.6 Data quality supportive supervision visit/audit
There has been one of the recommendations based on data audit in almost all rounds to conduct data
quality supportive supervision by the district team to support block facilities. The support by the
district/division/state team remained poor, however, it improved in the current round (35%)
compared to the last round (21%). Only one-third of facilities (35%) have been supported by any data
quality visit.
D. Source documents availability for the recording of data elements
The availability of registers with the provision of recording of information is the base for reporting
accurate information every month. Besides the audit of 61 data elements, the recording provision of
4 critical domains (ANC, Delivery, Family Planning, Child Health) with the availability of different types
of registers in the facility were also assessed during the audit (Table 7 & 8). The average number of
registers remained almost the same at DH and CHC over the sixth, seventh & eighth rounds, however,
high variations have been observed across same type of facilities in the last three rounds across all the

17
domains. Also, about half of the registers available at the facility were prepared manually and
frequently used across all domains. The average number of source documents by facility type and by
type of register over the last three rounds are given in Tables 8 & 9.
Table 7: Source document availability over the last three rounds by type of facility
Average number of source documents (printed & manual both) (Min-
Max) (N=24)
Domain
DH CHC
R-6 R-7 R-8 R-6 R-7 R-8
Ante natal care 4(1-4) 4(2-5) 4(2-5) 3(0-6) 3(1-5) 3(1-4)
Delivery/Newborn
care & 9(4-12) 10(8-13) 11(7-13) 7(3-12) 9(5-13) 7(4-11)
complications
Family planning 8(5-11) 6(4-8) 7(3-9) 6(0-11) 5(2-8) 6(2-8)
Child health 3(0-3) 2(0-3) 1(0-3) 2(0-3) 2(0-3) 1(0-3)

Table 8: Source document availability over the last three rounds by type of register
Average number of source documents (printed & manual both and
manual only) (Min-Max) (N=24)
Domain
All (Printed & manual) Manual only
R-6 R-7 R-8 R-6 R-7 R-8
Ante natal care 3(0-8) 3 (1-5) 3(1-5) 2(0-7) 1(0-4) 2(0-4)
Delivery/Newborn
care & 8(3-12) 9(5-13) 8(4-13) 2(0-7) 4(0-10) 4(0-8)
complications
Family planning 7(0-10) 6(2-8) 6(2-9) 1(0-10) 1(0-4) 1(0-6)
Child health 2(0-3) 2(0-3) 1(0-3) 1(0-3) 1(0-3) 1(0-2)

Four major domains are captured through 217 data elements from the monthly reporting formats of
HMIS and UPHMIS. The recording of these data elements has been assessed by observing the
presence of source documents only. Data captured by type of facility is given in Table 9. There has
been considerable improvement in the capture of data elements through records from 31% during the
sixth round to 54% during the seventh round and it sustained till the eighth round. This improvement
has been across all four domains. More than half (54%) of the data elements are currently recorded
by the audited facilities. This ranges from 34% of child health (out of 80 data elements) to 74% of
delivery and complication (out of 64 data elements) related information during the eighth round. Still,
there are a considerable number of data elements across different domains to be captured.

18
Table 9: Data elements captured at the facility
% of data elements recorded (N=24)
Domain DH CHC Total
(# of data elements)
R-6 R-7 R-8 R-6 R-7 R-8 R-6 R-7 R-8
Ante natal care
29 66 62 32 52 49 31 56 54
(37)
Delivery/Newborn
care & 49 80 80 37 61 72 41 67 74
complications (64)
Family planning
39 81 56 40 58 52 40 65 53
(36)
Child health (80) 17 36 45 11 23 31 13 27 34
Total 34 66 60 30 49 51 31 54 54

The facility wise gaps and action plan are annexed as Annexure 2.

6. MAJOR CHALLENGES

The challeges found in eight round of audit were almost same as observed during the last round.
However, in the recent time, the program priority of district, block and DH facilities shifted towards
Covid-19 management since the last round of audit conducted in Feb 2020. The services of
DWH/DCH were less affected as compared to CHCs during COVID. It not only hampered the
record keeping at CHCs but also the TSU Nurse mentors moved from CHC to DH during covid
which might be resulting into a dip in data matching. During this period, supportive supervision visits
could not be routinely planned by division and state officials which also resulted into the marginal dip
of the data matching levels from the previous rounds.

In addition, following challenges were observed by the team during data audit:

a. Non-functional validation committee meeting: The validation committee was constituted

to validate the reported data and ensure the quality of reported data at the block, DH & district
levels. Facilities during the eighth round have shown a fall in validation committee meetings;
it was observed that one third (35%) of the visited facilities (block facility and DH) did not
conduct validation meetings during the last quarter compared to one fifth (21%) during the
previous audit. The majority of the blocks where validation meeting took place, it was not as
per the guideline. The following issues were observed while interacting with the committee
members:
i. Validation committee meetings were not conducted regularly as per norms.
ii. No focus on key data elements related to ranking, NITI AYOG and hospital
performance in the meeting
19
 iii. No clear action plan was not developed and followed up.

b. Understanding issue with some data elements: While data audit it came to notice that there
is an understanding issue for the reporting ANC data elements specially 4ANC & 4HB, and
data elements on maternal & newborn complications and full immunization. Training of service
delivery staff may be a major reason as two-thirds of SNs & ANMs are yet to receive training
on HMIS & UPHMIS format reporting.

c. Absence of preparation of monthly summary in a register: HMIS and UPHMIS are the
two monthly reporting portals which require a monthly compilation of information from the
source documents. Child health, OPD and IPD were the domains where a monthly summary
was not prepared at the majority of the facility. However, it was also observed that monthly
summary preparation was usually less across most of the domains. The absence of a monthly
summary leads to wrong or blank reporting of the services provided by the facilities.

d. Data element-wise accountability of staff is missing: There are 317 data elements in HMIS
format now and almost the same in UPHMIS format. There are many reporting points in a
facility (PHC/CHC/DH). To ensure complete reporting all staff are supposed to share the
information such as LT to share lab information, SN to share delivery & newborn related data
elements, MO to share OPD related details and so on. However, many of the staff are not
aware of regarding reporting. Therefore, the completeness of the format is affected.

e. Non-uniform and non-availability of source documents (registers): Correct and optimal

recording of individual information in the register is the base for any reporting. The correct
recording involves the availability of source documents and having a provision to record all the
information supposed to be reported without any duplication. The non-uniform and
unavailability of source documents were observed as the major bottleneck for reporting quality
data. There was no provision for recording around half (46%) of the data elements (in four
major domains) of HMIS/UPHMIS which were supposed to be reported by the facilities. This
varied significantly for different domains and facilities but the overall level remained low across
all the domains (34% of recording provision in child health to 54% in ante-natal care) during
the eighth round of data audit also.

Besides this, a huge disparity in the available number of registers was also observed among
different facilities. Also, about 50% of the registers were manually prepared by facility staff
which had duplicate information and added a burden to the data capturing.

20
7. SUGGESTIVE SOLUTIONS

a. State-level data quality review meeting with divisional M&E hub

The validation committee was constituted to validate the reported data and ensure the quality of
reported data. The state office has also issued a guideline to conduct the meeting at block, district
and DH.
A quarterly state-level data quality review meeting can be a good platform to review the data quality
of the state with divisional M&E hub and their accountability can also be established.

b. Data element-wise accountability of staff

Data element-wise accountability of staff need to be fixed and verified by MOIC at the block and
by CMS at the DH. A suggestive in-charge & source document for all data fields in HMIS &
UPHMIS monthly reporting format at DH & CHC has been attached as Annexure-3.

c. Scale-up of data quality audit (data quality supportive supervision) at division level by
divisional M&E hub
It is important to have supportive supervision visits of the districts by divisional M&E for
continuous improvement in data quality. The divisional M&E officer must build the capacity of
district (DPM/DDM/HMIS operator/Hospital Manager) and block-level staff
(BPM/DEO/Nurse mentor) Ato analyze and report quality data. It is equally important to
prioritise the facilities/blocks by the divisional M&E officers. This prioritization can be based on
the identified gaps through data analysis. The continuous support by the divisional M&E hub will
also strengthen the validation committee meeting at the district and block levels.

d. Monthly summary of reporting data elements in the record

All reporting staff must prepare a monthly summary for reporting data elements on the record.
It shows accountability for reporting data elements.

e. Standardization of source documents

The availability of non-uniform registers causes a lot of burden on facility staff which further leads
to duplication of their efforts too. There is a need to review the available registers and recommend
a standard register to fulfil all the program need based on the findings over the last three rounds.

f. Promoting enahace data use: Increased data use leads to improved data quality. The district
must be skilled to analyse the data available in the district for tracking the progress of different
health schemes/programs for corrective actions on time. NHM M&E officers and TSU M&E
specialist at division can capacitate the district and block for the data anlaysis which may lead
correct reporting and data use in the district.

21
GLIMPSES OF DATA AUDIT

Picture 1: Feedback meeting on data quality audit at CMO office, Picture 2: Data Quality Audit at CHC Ratsar, Garwar Block, Ballia
Jalaun

Picture 3: Feedback meeting on data quality findings with CMS, Picture 4: Data quality audit at CHC Kara, Kaushambi
Sambhal

22