Facilitators, best practices and barriers to integrating family planning data in Uganda’s health management information system

The HMIS in Uganda is a standardized and integrated national reporting system. Standardized electronic and paper-based forms are used. Although the MOH is encouraging electronic-based data collection and reporting at all health facilities, the norm in private facilities and some lower-level and rural facilities is still to use paper-based forms due to shortages of computers, power, and internet connectivity. All health facilities (i.e., public, private for-profit, and private not-for-profit) are required to use the MOH’s standardized forms to avoid duplication of data and reduce overload of reporting to the HMIS. Records from community health workers are captured at the nearest health facility. The system requires adherence to correct procedures for compiling data and is designed for continuous cross-checking of data to eliminate errors. An example of a standardized form is HMIS Form 105 which reports monthly attendance figures for maternal and child health and FP visits, diagnoses for the outpatient department, laboratory tests, HIV and AIDS service data, stock outs of essential drugs and supplies, and financial data [21]. At a lower-level health facility, a records officer observed the following:

Everything is incorporated within that user form. It is integrated and so FP data is catered for.

Similarly, a national-level stakeholder observed:

All facilities use standardized forms for reporting; the individualized and special reporting by health facilities was abolished. Thus, there is less confusion and less workload given the similar tool used in data collection at all levels.

Most HMIS personnel at the district level had access to computers. Implementing partners, such as The AIDS Support Organization (TASO) provided computers to their partner health centers in Jinja.

The introduction of DHIS 2 by the MOH, with provisions for web-based reporting, strengthened district -based and national-level reporting. District-level HMIS personnel found the DHIS 2 software appropriate and user-friendly. Web-based reporting makes sharing health data easier among stakeholders who have user rights and can access the system. It allows IPs and donors to monitor and scrutinize the quality of data being collected. A national-level key informant made the following observation concerning HMIS:

It is an online system that captures all the paper forms as electronic and data is sent electronically. It is a server and at the same time entry point. Biostatisticians run through the data to rectify errors. Some errors are rectified automatically but others are sent back to the facility. Web-based reporting is an innovation which does not require district HMIS officers to submit data to the ministry in hard copies, hence reducing on the transportation costs as well as minimize paper use.

A key informant from Hoima District noted the following:

It is one of the best data collection systems we have in the country. I have had experience overseeing Buhanguzi Health Sub-District, which has 26 facilities. The system is easily adopted by the officers, and reporting becomes easy. In particular, since 2013, there have been tremendous positive changes [in reporting].

The Government of Uganda prioritizes FP in line with its commitment to international and regional conventions (e.g., International Conference on Population and Development 1994 and FP2020). Owing to this commitment, there is substantial emphasis on improving FP service provision and consequently collecting FP data. In addition, FP is among the country’s priority areas since the contraceptive prevalence rate is monitored at the national level.

In order to ensure compliance with the HMIS, the MOH has linked submission of HMIS reports with license renewal for private health facilities. All health facilities are required to regularly submit HMIS reports to the district HMIS office. For private facilities that are not reporting consistently or fail to report, a condition stipulates that prior to the renewal of their health facility license, they must submit all missing reports to the HMIS district office. For public health facilities, when submission of HMIS reports from any one district is delayed, the records assistant receives follow-up phone calls from the district HMIS focal person asking him or her to explain the delay. At the end of the year, penalties are imposed upon districts with very poor performance.

Availability of HMIS forms in registered and licensed health facilities (i.e., private, NGO, public) facilitates the HMIS in Uganda. These forms must be submitted to the districts and, thereafter, to the MOH. FP data from community outreach programmes by private or NGO providers are captured using government health facility registers. A national-level informant noted the following concerning public facilities:

The forms are available especially in the public health facilities. You find forms in all health facilities. Even recently, in the remotest health facility in the Karamoja Region, they had the HMIS forms.

In addition, continuous review of the HMIS forms has been a favorable condition. Every 5 years, the MOH invites all HMIS stakeholders (e.g., UNFPA, PACE, Marie Stopes Uganda) and health unit staff (e.g., biostatisticians, records assistants) for a meeting to provide input on improving the HMIS forms.

Training designated staff in HMIS data management is a critical facilitating factor. While Hoima District reported universal training coverage, Jinja’s training coverage was reported to range from 50 to 90%. Many officers reported receiving training in HMIS data management several times (as recent as the year of data collection). Trainings address computer literacy, navigating DHIS 2, data analysis, and updates on the contents of the form. In Hoima District, all health staff attending to outpatients had been trained on HMIS data collection, including FP. A biostatistician pointed out:

There is improvement in quality. I am not saying that [it] is very good data, but there is improvement in quality because of this regular training and introduction of this web-based reporting. I think these contribute to quality because before they would collect all the data and bring to the ministry to enter and you know what it means—and now sometimes the data entry is done at the facility.

Supportive supervision is among the key components of the HMIS. Health facilities receive quarterly visits, mainly from district officials, and supportive visits from MOH officials. IPs, such as the World Health Organization, the United States Centers for Disease Control and Prevention, and World Vision support and participate in supervision and trainings and are usually available for consultation. Facility-level records officers receive supervisory visits from district HMIS focal persons and organizations and donors such as the United States Agency for International Development (USAID), World Health Organization, World Vision, PACE, TASO, and Infectious Diseases Institute (IDI). The MOH, through the Regional Performance Monitoring Team, conducts quarterly visits to the districts and occasionally to health facilities. Visits are followed by review meetings on how to improve service delivery. The team supports networks and linkages between IPs and health facilities.

In general, the HMIS was reported as having adequate human resources, facilities, software, and forms. In addition to government and USAID support, districts mobilize resources through collaboration with IPs. Such IPs support capacity building, training, service delivery, and commodity provision. Examples of IPs associated with FP programs in Uganda are Reproductive Health Uganda, Marie Stopes Uganda, World Vision (which also focuses on maternal, neonatal, and child health), and IDI. Monitoring and Evaluation Technology Support, a project based at the Makerere University School of Public Health that is interested in data, supports printing DHIS 2 forms for all health facilities and monitoring all data-related issues. TASO receives data from the 17 higher-level health facilities it supports. TASO staff receive free data bundles from Mobile Telecommunications Network that facilitate access to the MOH website.

A high level of appreciation and motivation of data users and managers is a key facilitator for enhancing HMIS performance in Uganda. IPs provided financial incentives to biostatisticians and HMIS focal persons in both Hoima and Jinja Districts. Users and managers rated the HMIS as one of the best data collection and health management systems covering different health aspects. All respondents except one were motivated to work with the system because of their interest in data analysis (if granted access), data management experience, and desire for high-quality data for planning purposes. In support of the system, a national-level officer from an NGO observed:

No work documented is no work done. People should be encouraged to document what they have exactly done.

Most respondents noted that health personnel were generally competent and had been cooperative in executing HMIS tasks. Relevant personnel exhibited flexibility and willingness to adapt new methods and the addenda included in the HMIS forms. Since training programs entail assessments, personnel work hard to protect their names and positions since they value their jobs. In addition, team work is promoted among the HMIS personnel, which contributes substantially to the quality of outcomes. A national-level and district-level respondent, respectively, said:

Team work is key. Where staff work together in generation, entry and discussion of the reports has yielded quality data. In some districts, district health teams meet every end of the month to check data quality.

We do our work then at the end of the month, we sit on a round table. We compile, analyze, and then submit. If there are any errors, again we talk about them and you know. It is not a one man’s business.

Collaboration or networking with other districts or actors is beneficial to the process of data management. Partnerships have made work easier and contributed to improvements in meeting reporting timelines.

With respect to consistent use of HMIS forms, national informants noted that public health facilities are more consistent in health information reporting because staff’s access to the forms is guaranteed and ensured. Unlike private facilities, public facilities receive regular supervision. An MOH informant observed that low coverage of HMIS forms at private facilities is attributed to fewer trained staff at those facilities:

In the private sector, reporting is less than 25%. Apart from recording the name and date of review or re-attendance, nothing much is done in the private facilities. Staff need to be brought on board, to be trained and then mentored on how useful it is to capture data.

Health facilities receive reminders from the district office when reports are due. All health facility records staff (from both private and public facilities) are expected to prepare HMIS reports for submission as stipulated in the HMIS health unit procedure manual [3]. In this manual, each health unit is expected to keep a register of all patients (Form 031) that is updated daily. From the register, tally sheets are used to populate the HMIS reporting forms. A records officer elaborated on the process as follows:

Data are captured on a daily basis at the health facilities using the 031 form. Weekly, we report on surveillance, and that is Form 033B. We also have a monthly report and this is usually on the general condition in the facility like drugs, store medical, theatre, etc., and this is HMIS 105 for outpatient and 108 for the inpatient. Quarterly, we compare three months’ information and it’s done manually. Then we have the annual report that is HMIS 107 where we generate all months and come up with one report. Finally, we have the financial year report.

Reporting is done on a weekly basis using mobile tracing (mTrac)—a health system strengthening tool using text messaging. Data reported via text through mTrac are immediately made available within DHIS 2 for further analysis. A parallel system entails submitting hard copies of the tally sheets to the facility HMIS focal person. A provider prepares a monthly report and submits it to the district HMIS officer by the seventh of the following month for compilation and entry in DHIS 2. Monthly reports contain detailed health information from all health facilities. The district HMIS office submits all reports electronically to the MOH Resource Centre by the twenty-eighth of the following month. The MOH Resource Center disaggregates data by district, health center, and type of health facility (i.e., private or public).

Standardized and convenient data transmission procedures are used. Personnel involved in data transmission are trained in reporting procedures. Furthermore, the online reporting form makes it easy to detect mistakes. The process has been simplified with the mTrac system enabling mobile phone users to send data for weekly reports. A records assistant from Hoima reported:

We used report on a weekly basis but right now we no longer do so because we are provided with the mTrac system. So, we can use our phones to send data. Now we send weekly reports using [the] mTrac system using Internet on our phones.

Health facility in-charges are advised to allow records personnel to submit data directly to the district since the records personnel must receive feedback from the district on their submissions. Data in the national database can be easily retrieved and used for decision making.

Feedback at the various levels is both bottom-up and top-down. Districts provide supportive supervision to health facilities and advise staff on areas for improvement. District teams help health facilities address challenges and facilitate a process in which the worst-performing health facilities can learn from the best-performing ones. The DHIS and IPs communicate concerns about data accuracy and reporting time. Reports are graded and performance is displayed, which is a motivator for better performance. IPs also regularly share information, provide feedback, and suggest recommendations for improvement. This was particularly the case in Jinja District. The MOH provides feedback to districts on the quality of reports, data cleaning issues, reporting rates, timeliness, system issues, and performance. Coaching visits are conducted and, in such sessions, district personnel coach and obtain feedback from health facility personnel, and vice versa.

Quarterly performance reviews are conducted at the national level to provide feedback to relevant personnel. These meetings ensure that data collection and reporting is in line with HMIS data collection and reporting procedures. During performance review meetings, district HMIS focal persons and health management team members review the performance of each health facility. They then request staff from facilities that are performing well in terms of timely reporting to share with other facilities to facilitate adoption of good practices. Similarly, staff from facilities that are not performing well are requested to share their experiences, why their performance is not up to standard, and how they can be assisted. One stakeholder from Jinja District explained:

We normally conduct performance review meetings quarterly. We look at how facilities have performed in different areas. Sometimes we spot out the best-performing facilities and the worst-performing facilities and we look at how the best ones have been managed and then we look at the worst-performing ones and what has caused them not to perform so that they tell us because of: say, “It is hard to reach,” “We are at the island so while trying to reach the place the boat got a problem and we took two weeks.” We share all that in review meetings.

Collaboration between the MOH and IPs has been key. Although the MOH oversees HMIS implementation within the districts, it collaborates closely with IPs that support HMIS work through capacity building, training, service delivery, supervision, and commodity provision. Such IPs include USAID, the Centers for Disease Control and Prevention, World Health Organization, Marie Stopes Uganda, PACE, and IDI. Companies, such as the Mobile Telecommunications Network, support the HMIS by providing free data bundles to HMIS staff. This facilitates communication and results in faster access to MOH websites and improved use of DHIS 2 software. According to a national-level information officer and a records officer, respectively:

These partnerships have “made life easy.” It is easy to address challenges, there is improvement in timelines in reporting. IPs also offer financial support.

Based on the facilitators, we identified the following best practices for integrating public and private data into one HMIS. These best practices pertain to not only FP data, but all health data:

The complexity of the HMIS reporting form and procedures is challenging. Respondents noted that the reporting form is very long (too many pages and indicators), lacks adequate space for capturing data and does not have a place to indicate the sex of the FP user. Some codes used for FP data on the form are not available in Ugandan health facilities. The form does not provide explanations on meaning of the codes. Furthermore, some of the contents are not applicable to lower-level facilities. A district level informant intimated:

The reporting form is not user-friendly. It has many pages with lots of information requirements. This makes it hectic. So, often times staff submit incomplete and inaccurate forms.

The paper-based reporting forms (often used in rural facilities) and the online platforms have not been completely harmonized. Some of the data captured on the paper-based HMIS forms are not incorporated in the online DHIS 2. In addition, records personnel are sometimes asked to collect and capture information that is not on the paper-based form. This leads to submission of incomplete and inaccurate forms. The volume of the tool and manual is huge, which is a disincentive to staff.

Although regular revision of forms is good practice, it takes time for records personnel to become familiar with the revised forms. The implication is that if a facility uses an older version of the tool, some information is missed or not filled in.

Although two-step data entry could have value with respect to quality control, double entry of HMIS data makes the procedures strenuous. Records assistants are required to fill out hard copies of the forms that are later submitted to the district biostatistician for entry in DHIS 2. It would take less time if data were entered directly in the system and thereafter verified by the biostatistician or HMIS focal person. Some of the records officers reported that they were trained in information systems and can handle the online system.

Besides the lengthy form, some health facility personnel found the DHIS 2 software slow in generating reports. The form is designed in such a way that it does not upload when there are missing data.

Despite the availability of computers at the district level, some lower-level facilities at the sub-district level still lack computers. This presents a challenge for data entry and verification. The lack of computers increases the workload of the district biostatistician and HMIS focal person. Most facilities use traditional client registers to extract data and populate the HMIS reporting forms, which are later submitted to the district HMIS focal person. At the district level, most of the computers have old versions of computer software, lack antivirus protection, and are poorly maintained.

There are problems with Internet connectivity, which delays the transmission of data to the MOH HMIS dashboard. Similarly, inconsistent power supply at the district level delays data transmission to the MOH servers. Not all facility- or district-level staff have access to the MOH dashboard. Sometimes personnel that can access to the system and dashboard (e.g., IPs, district officers) are busy and cannot give information when it is required.

Staffing for the HMIS is inadequate. A senior biostatistician from the MOH reported that only five of the 15 national-level staff are available to supervise the HMIS. This is a limited number given the workload and the effort needed for effective implementation of the system. At the district level, the biostatistician is expected to capture data from all health facilities in the HMIS and perform data assessment, analysis, and validation. This heavy workload affects the timeliness, completeness, and quality of the district reports. In some health facilities, only one nurse provides FP services. Shortages in human resources result in late and under-reporting of FP data. One district-level informant said:

I am the only person in this office who captures data from all the health facilities into the system, does data assessment, analyzes, validates, and everything else.

Public facilities usually have fewer staff. Therefore, because of the heavy workload, staff on duty may not record FP services. Private facilities, on the other hand, may be required to report to other agencies in addition to the mandatory reporting at the district for the MOH, leading to reporting fatigue.

Constant training on the HMIS is required due to high staff turnover or attrition, especially in private facilities. A key informant from Hoima District reported that staff in private facilities frequently leave for better opportunities, resulting in recruitment of unqualified and inexperienced personnel to handle HMIS data. This creates the need to recruit and retrain new records personnel for such facilities, which is both costly and time-consuming.

When forms are revised, not everyone is trained on the changes, and this creates a barrier for data quality. Some records assistants had neither received adequate training nor refresher training. One district-level informant added:

I have never heard of nurses and midwives going for refresher training on family planning data in the HMIS.

Irregular training in the context of staff attrition in private facilities and revision of HMIS forms, negatively affects data quality in terms of accuracy and completeness at the district level, since responsible personnel have to learn on the job.

Several health facilities were reported to have stock outs of HMIS forms and stationery, which affects reporting. Sometimes stationery is not provided in time and districts have no budget for printing and photocopying materials for the facilities. Jinja District reportedly went without in-patient registers for a couple of years, and many facilities (both public and private) do not have FP registers. This leads to compilation of incomplete FP data since some records are made when services are provided to clients.

Registration and coding of facilities in the national HMIS by the MOH takes a long time. The district usually monitors and assesses a health facility before incorporating it in the HMIS despite commitments from IPs such as PACE. This contributes to incompleteness of data because the system is not covering all health facilities in a districts.

There is lack of support for FP data from management in some religious health facilities. For example, in some Catholic health facilities the FP section is left blank and is therefore recorded as incomplete.

There is a general lack of interest in and demand for FP data on the part of IPs. Limited attention is given to FP data at the health facility level, thus affecting data quality.

Heavy workloads, especially in private health facilities, which may not have designated records staff, have the potential for FP compromising data quality. Health workers who already burdened with heavy workloads, are also responsible for record keeping. In such contexts, record keeping and documentation of FP data is often deemed a secondary priority. Hence, opportunities for checking data quality are limited.

There are challenges in collecting data on FP services provided by village health teams during community outreaches. In some cases, the services are not always recorded by health facilities. In addition, some records assistants have difficulty interpreting the FP codes that are found in the HMIS forms, leading to submission of incorrect data or information. One national-level informant reiterated:

A case in point are entries related to condoms. So, one is required to fill in a number. However, the question is whether they are referring to packets or number of individual condoms taken by a person. So, the response expected is not clear and these data remain vague. Some records assistants focus on other indicators compared to FP, so they often have no statistics in the respective entries.

With respect to completeness of information, informants noted that overall, reporting is around 87% for public and private facilities together. The delays are sometimes caused by lack of resources for transporting the paper-based forms to the district office, or misplacement of the forms by the relevant district staff.

Too many reports are required by the MOH, leading to heavy workloads for the HMIS focal person and district biostatisticians. IPs also have too many data demands but are given priority at the district level since they support district programs.