Integrating DHIS2 with AMRIT
Integrating DHIS2 (District Health Information Software 2) with the AMRIT platform for reporting involves creating a seamless interface that allows data to flow from AMRIT service lines to DHIS2 while ensuring accuracy, security, and compliance with healthcare standards.
Integration Architecture
Data Flow - Data collected from various AMRIT modules (e.g., Inventory, HWC, Scheduler, MMU) will be exported to DHIS2 for centralized reporting.
Data Exchange Mechanism - Use Python scripts and DHIS2 REST APIs to push data.
Data Mapping - Map AMRIT’s data structures (e.g., beneficiaries, health programs, inventory) to DHIS2’s data elements, indicators, and organization units.
DHIS2 (District Health Information Software 2)
Introduction:
DHIS2 (District Health Information Software 2) is an open-source software platform designed to manage, analyze, and visualize health data, particularly in low- and middle-income countries. It is widely used in public health systems for data collection, reporting, and analysis, making it an essential tool in monitoring and improving healthcare services.
DHIS2 Download Link: https://dhis2.org/downloads/
Key Features of DHIS2:
Data Collection: Allows easy data entry from multiple sources like mobile devices, tablets, or desktop computers. It supports offline data entry, syncing once the internet is available.
Health Monitoring: Tracks a wide range of health indicators such as vaccination coverage, disease outbreaks, maternal health, etc.
Data Analysis: Offers powerful tools to generate custom reports, charts, maps, and dashboards. Users can analyze data in real-time to improve decision-making.
Customizability: DHIS2 can be tailored to specific health programs or needs by creating custom forms, datasets, and workflows.
Integration: Can integrate with other health systems for interoperability and exchange of health data across platforms.
System Requirements
Server Requirements:
Operating System: Linux (Ubuntu recommended), Windows, macOS
RAM: Minimum 8 GB (24 GB recommended for large installations)
Processor: 4-core CPU or higher
Disk Space: 100 GB (depending on data size)
Software Dependencies:
Java: Java SE 8 or higher (DHIS2 runs on Java Runtime Environment)
Database: PostgreSQL (Version 10 or higher)
Web Server: Apache Tomcat 8 or higher (for hosting DHIS2)
Others: Nginx (Optional, used for reverse proxy)
Installation Guide:
Install Java
Open the terminal and run the following command to install Java 8:
Verify Java installation:
Install PostgreSQL
Install PostgreSQL database server:
Create a new PostgreSQL role and database for DHIS2:
Set a password for the PostgreSQL user:
Edit postgresql.conf to enable remote connections if needed:
Change the following line:
Install Tomcat
https://tomcat.apache.org/download-90.cgi
Download and install Tomcat:
Increase the Tomcat memory allocation by editing the setenv.sh file:
Add the following lines to allocate memory (adjust as needed):
Deploy DHIS2 WAR File
Download the DHIS2 WAR file from the official DHIS2 site (https://releases.dhis2.org/ )
Move the WAR file to Tomcat's webapps directory:
Start Tomcat:
Access DHIS2 via your browser:
Configure DHIS2
After logging in, use the default credentials (
admin/district) to access the platform.Set up your organization structure by navigating to:
Create datasets and assign them to specific organizational units to start data entry and collection.
Using DHIS2
Data Entry
To enter data, go to:
Enter the required data and submit. This data can now be analyzed and visualized in reports.
Reports & Analysis
Reports:
Navigate to:
Select a report, specify the organization units and period, and generate the report.
Visualizations:
Go to:
Select data elements or indicators to create custom charts, tables, and graphs.
Maps
Use the GIS App to create maps showing health data geographically:
User Management
Creating Users
Navigate to:
Create users and assign roles based on the permissions required (e.g., data entry, report viewer).
Setting User Roles
Define user roles in:
Assign different levels of access (e.g., system admin, data input user).
Security Best Practices:
Always change the default admin password after installation.
Implement role-based access control (RBAC) to restrict access to sensitive data.
Regularly back up the PostgreSQL database using:
Enable HTTPS for secure communication by configuring SSL in Tomcat or using a reverse proxy like Nginx.
Maintenance & Backup:
Updating DHIS2
Download the latest DHIS2 WAR file from the official site.
Stop Tomcat:
Replace the old WAR file in
/var/lib/tomcat9/webapps/with the new one:
Start Tomcat again:
Database Backup
Regular backups are essential. Use the following command to back up the PostgreSQL database:
You can automate this by setting a cron job in Linux:
Add a backup job to run daily:
Troubleshooting:
Common Issues
Tomcat Not Starting:
Check if the correct Java version is installed.
Review Tomcat logs in
/var/log/tomcat9/for specific error messages.
Database Connection Failure:
Ensure PostgreSQL is running and accepting connections.
Verify that database credentials in the DHIS2 configuration file (
dhis.conf) are correct.
Logs and Monitoring
Monitor Tomcat logs:
Check PostgreSQL logs for database-related issues:
DHIS2 Architecture
DHIS2 Metadata creation: https://dhis2.org/
1. Introduction to Metadata in DHIS2:
Metadata in DHIS2 refers to the structural elements that define the data collection process and how data is organized within the system. The key metadata components are:
Organization Units: Geographic or administrative units such as regions, districts, or facilities.
Data Elements: The smallest unit of data, representing what is being collected (e.g., number of vaccinations).
Datasets: A collection of data elements used for data entry.
Indicators: Calculated values that give insight into performance or outcomes (e.g., vaccination coverage).
Categories & Category Options: Define different disaggregation’s (e.g., by age, gender).
2. Creating Organization Units:
Organization Units represent the hierarchical structure of the health system or organization (e.g., country > region > district > facility). These must be set up before collecting data.
Steps to Create an Organization Unit:
1. Navigate to the Maintenance App:
Go to
Apps>Maintenance.
2. Create Organization Units:
In the left menu, go to
Organization Units>Add.Enter the following details:
Name: Name of the organization unit (e.g., "District Hospital").
Code: A unique code (optional but recommended).
Parent Unit: Select the parent unit (e.g., if it's a district, its parent would be a region).
Opening Date: When this unit became operational.
Click Save.
3. Organization Unit Hierarchy:
You can create more organization units under the parent units to establish a hierarchy (e.g., country > state > district > health center).
3. Creating Data Elements:
Data Elements are the core items of data collected in DHIS2. For example, "Number of children vaccinated" or "Malaria cases reported."
Steps to Create a Data Element:
1. Go to Maintenance:
Navigate to
Apps>Maintenance.
2. Add a New Data Element:
In the left menu, go to
Data Elements>Add.Fill in the following details:
Name: Name of the data element (e.g., "Children Vaccinated").
Short Name: A shorter name for quick reference.
Code: Optional but useful for integration with other systems.
Value Type: Select the type of value this data will hold (e.g.,
Number,Text,Boolean).Aggregation Type: How the data should be aggregated (e.g.,
Sum,Average).Domain Type: Choose if it's
Aggregate(for summary data) orTracker(for individual data).Category Combination: Specify how the data is disaggregated (e.g., by age or gender).
Click Save
3. Creating Datasets:
Datasets define the collection of data elements that will be used for data entry on a regular basis (e.g., weekly, monthly). Each dataset is linked to specific organization units and time periods.
Steps to Create a Dataset:
1. Go to Maintenance:
Navigate to
Apps>Maintenance.
2. Add a New Dataset:
In the left menu, go to
Datasets>Add.Enter the following information:
Name: The name of the dataset (e.g., "Monthly Vaccination Data").
Period Type: Select the frequency of data collection (e.g.,
Monthly,Weekly,Daily).Category Combination: Define any disaggregation (e.g., age, sex).
Data Elements: Add the data elements that will be included in the dataset.
Assigned Organization Units: Specify which organization units are responsible for data entry (e.g., district health centers).
Click Save.
5. Creating Indicators:
Indicators represent calculations based on data elements to provide insights (e.g., "Immunization Coverage"). Indicators can represent percentages, rates, or ratios.
Steps to Create an Indicator:
1. Go to Maintenance:
Navigate to
Apps>Maintenance.
2. Add a New Indicator:
In the left menu, go to
Indicators>Add.Enter the following details:
Name: The name of the indicator (e.g., "Vaccination Coverage").
Short Name: A short name for easier display.
Numerator: Define the formula for the numerator, which is usually a data element (e.g., "Number of vaccinated children").
Denominator: Define the denominator, usually a population data element (e.g., "Total children population").
Indicator Type: Select the type (e.g.,
Percentage,Number).Click Save.
6. Creating Categories and Category Options
Categories allow you to disaggregate data (e.g., by age, sex). For example, you may want to record "Number of Vaccinations" by both "Age Group" and "Gender."
Steps to Create Categories & Category Options:
1. Go to Maintenance:
Navigate to
Apps>Maintenance.
2. Create Category Options:
In the left menu, go to
Category Options>Add.Example: Add options like "Male," "Female" for gender, or "Under 5," "Above 5" for age groups.
Click Save.
3. Create Categories:
In the left menu, go to
Categories>Add.Define a category (e.g., "Age Group").
Assign the category options (e.g., "Under 5," "Above 5").
Click Save.
4. Create Category Combinations:
In the left menu, go to
Category Combinations>Add.Combine multiple categories (e.g., "Age Group" and "Gender").
Click Save.
7. Assigning Metadata to Organization Units:
Once the metadata is created, you need to assign it to the relevant organization units to ensure that they can use the data elements, datasets, and indicators during data entry.
Steps to Assign Metadata:
1. Go to Maintenance:
Navigate to
Apps>Maintenance.
2. Assign Datasets to Organization Units:
In the
Datasetssection, open a dataset and select the organization units that will report using this dataset.
3. Assign Data Elements to Categories:
In the
Data Elementssection, assign the appropriate category combinations (e.g., Age, Gender) to each data element.
8. Metadata Export and Import:
For large implementations, you may want to export metadata from one instance of DHIS2 and import it into another.
Steps to Export Metadata:
1. Go to Import/Export App:
Navigate to
Apps>Import/Export.
2. Export Metadata:
Select
Export Metadata.Choose which metadata (e.g., data elements, datasets) to export.
Download the file.
Steps to Import Metadata:
1. Go to Import/Export App:
Navigate to
Apps>Import/Export.
2. Import Metadata:
Select
Import Metadata.Upload the metadata file exported earlier.
Check for conflicts and import.
DHIS2 Hosting
Hosting can be done in on-premise servers or in the cloud. DHIS2 is open source software and can be installed at your servers and used for free. You can also go for a professionally managed DHIS2 instance in the cloud. A managed DHIS2 instance takes care of the backup, security, monitoring and high-speed connectivity aspects of the deployment and allows you to focus on the information system itself.
PostgreSQL Examples
DHIS2: Demographic data:
Based on the beneficiary registration information first we need to create a data structure (metadata) and then import the required beneficiary demographic data from AMRIT Database to DHIS2 platform using either excel or through Python script we can integrate the data through API.
Query:
SELECT mb.BeneficiaryID,i_ben_mapping.BenRegId AS mappingBenRegId, i_ben_details.BeneficiaryRegID,
i_ben_details.FirstName,i_ben_details.MiddleName,i_ben_details.LastName,i_ben_details.Gender,
YEAR(i_ben_details.DOB),
IF(i_ben_details.DOB IS NOT NULL,DATE_FORMAT(i_ben_details.DOB, '%Y-01-01'), NULL) AS DOB,
i_ben_mapping.VanID,vanmaster.VanName,f.FacilityName,i_ben_mapping.CreatedDate,CAST(i_ben_mapping.CreatedDate AS DATE),
i_ben_address.PermState,i_ben_address.PermStateId,i_ben_address.PermDistrict,
i_ben_address.PermDistrictId,i_ben_address.PermSubDistrictId
FROM sehatok_db_identity.i_beneficiarymapping i_ben_mapping
INNER join sehatok_db_identity.m_beneficiaryregidmapping mb on mb.BenRegId=i_ben_mapping.BenRegId
INNER JOIN sehatok_db_identity.i_beneficiarydetails i_ben_details ON
i_ben_details.BeneficiaryDetailsId = i_ben_mapping.BenDetailsId
INNER JOIN sehatok_db_identity.i_beneficiaryaddress i_ben_address ON
i_ben_address.BenAddressID = i_ben_mapping.BenAddressId
INNER JOIN sehatok_db_iemr.m_van vanmaster ON vanmaster.VanID = i_ben_mapping.VanID
INNER JOIN sehatok_db_iemr.m_facility f on vanmaster.FacilityID=f.FacilityID
WHERE i_ben_mapping.BenRegId IS NOT NULL
AND i_ben_address.PermStateId IS NOT NULL
AND i_ben_address.PermDistrictId IS NOT NULL
and vanmaster.ProviderServiceMapID = 3;
Transaction data/ Event data Importing:
Based on the Beneficiary registration we can import the event data in DHIS2 based on specific period-wise:
SELECT DISTINCT
t.BenCallID,
t.CallID,
t.BeneficiaryRegID,
t.CallTime,
t.CreatedDate,
-- t.CZcallDuration As CallDuration,
TIME_TO_SEC(TIMEDIFF(t.CallEndTime, t.CallTime)) AS CallDurationInSeconds,
CONCAT( UPPER(SUBSTRING(trim(concat(trim(rmu.UserName),"-",t.CallReceivedUserID)), 1, 1)), LOWER(SUBSTRING(trim(concat(trim(rmu.UserName),"-",t.CallReceivedUserID)), 2))) As Received_username,
if(emu.UserName IS NULL , 'Not Assigned-99999',CONCAT( UPPER(SUBSTRING(trim(concat(trim(emu.UserName),"-",t.CallEndUserID)), 1, 1)), LOWER(SUBSTRING(trim(concat(trim(emu.UserName),"-",t.CallEndUserID)), 2))) ) As End_username,
t.ReceivedAgentID AS AgentID,
t.ReceivedRoleName,
t.IsCalledEarlier,
t.IsOutbound,
mct.CallGroupType,
mct.CallType,
dt.Call_Status,
dt.SessionID,
dt.Agent_Disposition_Category,
dt.Agent_Disposition,
dt.DID_Number,
dt.Queue_time,
dt.Wrapup_time,
dt.Wait_Time,
dt.Actual_Talk_Time,
dt.Channel,
dt.Wrapped_By
FROM
db_iemr.t_detailedcallreport dt left join
db_iemr.t_bencall_vtbl t on (t.callid=dt.SessionID and t.ReceivedAgentID=AgentID)
LEFT JOIN db_iemr.m_calltype mct ON mct.CallTypeID = t.CallTypeID
LEFT JOIN db_iemr.m_user_vtbl rmu on rmu.UserID=t.CallReceivedUserID
left join db_iemr.m_user_vtbl emu on emu.UserID=t.CallEndUserID
WHERE t.BeneficiaryRegID IS NOT NULL AND t.IsOutbound IS NOT TRUE AND t.calltypeid IS NOT NULL
AND (dt.Call_Start_Time between '2024-09-01 00:00:00' and '2024-09-17 23:59:59')
Anonymization:
Anonymization in a database refers to the process of removing or altering personally identifiable identifiable information (PII) so that individuals can’t be identified from the data.
This is crucial for privacy and regulatory compliance, especially dealing with sensitive information sucha as personal details, health information, and financial information.
Masking
Masking involves replacing sensitive data with a modified version that still looks valid but has no real-world significance. This is often used in systems where data format need to be preserved but the information should not be identificable.
Example:
SELECT * FROM db_identity.i_beneficiarydetails;
create view i_beneficiarydetails_vtbl as
select `BeneficiaryDetailsId`, `BeneficiaryRegID`, `TitleId`,`Title`,
CONCAT(SUBSTR(FirstName, 1, 1), REPEAT('*', CHAR_LENGTH(FirstName) - 1)) as `FirstName`,
CONCAT(SUBSTR(MiddleName, 1, 1), REPEAT('*', CHAR_LENGTH(MiddleName) - 1)) as `MiddleName`,
CONCAT(SUBSTR(LastName, 1, 1), REPEAT('*', CHAR_LENGTH(LastName) - 1)) as `LastName`,
`Status`,`GenderId`,`Gender`,`MaritalStatusId`,`MaritalStatus`,`MarriageDate`,`MarriageAge`,
CONCAT(year(dob),'-00-00 00:00:00') as `DOB`,
CONCAT(SUBSTR(FatherName, 1, 1), REPEAT('*', CHAR_LENGTH(FatherName) - 1)) as `FatherName`,
CONCAT(SUBSTR(MotherName, 1, 1), REPEAT('*', CHAR_LENGTH(MotherName) - 1)) as `MotherName`,
CONCAT(SUBSTR(SpouseName, 1, 1), REPEAT('*', CHAR_LENGTH(SpouseName) - 1)) as `SpouseName`,
`EmergencyRegistration`,`HealthCareWorkerId`,`HealthCareWorker`,`PlaceOfWork`,`LiteracyStatus`,
`educationId`,`education`,`occupationId`,`occupation`,`incomeStatusId`,`incomeStatus`,
`communityId`,`community`,`religionId`,`religion`,`preferredLanguageId`,`preferredLanguage`,
`sourceOfInfo`,`servicePointId`,`areaId`,`zoneId`,`phcId`,`Remarks`,`familyid`,`HeadofFamily_RelationID`,
`HeadofFamily_Relation`,`Others`,`Deleted`,`Processed`,`CreatedBy`,`CreatedDate`,`Reserved`,`ReservedFor`,
`ReservedOn`,`ReservedById`,`ModifiedBy`,`LastModDate`,`VanSerialNo`,`VanID`,`VehicalNo`,`ParkingPlaceID`,
`SexualOrientationID`,`SexualOrientationType`,`IsHIVPositive`,`HIVStatus`,`SyncedBy`,`SyncedDate`,
`ReservedForChange`,`MonthlyFamilyIncome` from `db_identity`.`i_beneficiarydetails`;
Data Integration Process from AMRIT to DHIS2:
Visualizations:
After importing the data into DHIS2 application, will run the analytics and create the visualizations based on the requirement.
Dashboard Sharing:
After completion of the dashboard will assign the users to the dashboard and provide the required grant access to the user.
A public dashboard URL is also available to share the dashboards publicly anyone has access to the dashboards without any credentials.
Example: https://dashboards.piramalswasthya.org/dhis-web-public-dashboard/hihlp.html
Dashboard PDF format: Dashboard can print using Dashboard layout under more options.
Non-PII Data Extraction: DHIS2 reports Data can be extracted in the form of Excel from the Reporting Database and print the report in pdf format.
MMU/TMC projects:
Here we have both Print and downloading options available, downloading into excel format and printing in PDF format.
If we want to print the report we can select the print option in the standard reports, it will print into pdf format.
Request Payload Format
Payload {startDate: "2024-11-18", endDate: "2024-11-19", Program: " Mobile Medical Unit (MMU)", OU:”BSL MMU”,…}
Organisation Unit: BSL MMU
Program: Mobile Medical Unit (MMU)
fileName: "Beneficiary wise report"
Start date: “2024-11-18”
End date: “2024-11-19”
104 District Report:
Request Payload Format
Payload {startDate: "2024-11-01", endDate: "2024-11-01", Program: "104 service", OU:”Assam”,…}
Organisation Unit: Assam
Program: 104 service
fileName: "104 district report"
Start date: “2024-11-01”
End date: “2024-11-01”
1097 National Aids helpline:
Request Payload Format
Payload {startDate: "2024-11-18", endDate: "2024-11-19", Program: "1097 National Aids helpline", OU:”BSL MMU”,…}
Organisation Unit: India
Program: 1097 National Aids helpline
fileName: "Beneficiary wise report"
Start date: “2024-11-18”
End date: “2024-11-19”
References
Document Reference: https://docs.dhis2.org/en/home.html
Last updated
