Journal of Health Informatics in Africa

6

Authors: Rogers Kagimu, Kiberu Micheal Vincent

Background The advent of effective infection control and treatment measures has shifted the burden of disease from infectious causes to chronic; non-communicable diseases (NCDs). Diseases such as cardiovascular and cancer are on the rise globally with a rapidly growing crisis in low-income and middle-income countries (LMICs). In 2015 for example, cancer was responsible for 8.8 million deaths globally and approximately 70% of these deaths occurred in LMICs. With lung, female breast, stomach and colorectal cancers combined with liver and cervical cancer (CaCx) explaining over half the incidence burden (54%) in LMICs. For CaCx, the pattern is striking since around 85% of women diagnosed and 87% of those who die from the disease live in LMICs. In Africa, CaCx is high in the sub-Saharan region with Uganda being ranked 10th with 44.4|100,000 women; Age-standardized incidence rate of CaCx cases. Among the contributing factors for the high incidence of the disease in the country include; lack of screening tests and vaccines to prevent HPV limited access to cervical screening programs as well as lack of information among others. This study investigated the information needs priorities about CaCx and coping information access strategies for CaCx clients and the general public. Material The study was conducted at Uganda Cancer Institute (UCI) gynae-clinic and participants were women with CaCx (CaCx clients), women without CaCx (general public) and healthcare professionals in active care for CaCx (gynae oncologists, nurses, & clinical officers). All participants were first informed of the study and those willing to take part, were recruited between April and May, 2018. For eligibility, participants had to be adults of above 18 years, able to speak the local language (Luganda) or English or both, confirmed evidence of CaCx or CaCx negative. Clients who were attending the clinic for the first time, and unaware of their diagnoses were excluded. While recruiting CaCx clients, a list was generated and participants randomly selected. For healthcare professionals and women without CaCx, convenient sampling and simple random sampling were used. Interviews were audio recorded and hand written notes were done throughout the interview session. Focus Group Discussions (FGDs) comprised of participants with and without CaCx. Key informant interviews (KIIs); nurses (n=2) clinical officers (n=2) and gynae-oncologists (n=2), and In-depth individual interviews (IDIs) comprised of women without CaCx (n=4). Data collected was transferred from the recorders onto a secured password protected computer. Notes taken during interviews together with consent documents were securely kept under lock and key for further reference. Interviews were transcribed then verbatim transcriptions were generated. Analysis of data was done using Atlas.ti version7 taking an exploratory approach (inductive content analysis), and findings presented as themes. Research and Ethics Committees of Makerere University, and Uganda Cancer Institute (UCI-REC) approved this study, and participants consented to take part. Each participant consented before taking part in the study. Results It was evident that clients desire more extensive information concerning the disease, treatment, and prognosis. Important information themes are highlighted in bold italics. CaCx Symptoms CaCx noted that if information on symptoms was known to them and to the general public, there is/was a high likelihood to seek medical attention when the disease has not reached advanced stages. Screening Participants needed screening information on where to find this service, how it is done and who to seek for it and when. Such information will help in early diagnosis; a point when medical intervention can still make a long-term difference. Dos and Don’ts Information concerning health-lifestyle which contributes to improved health was among what needed to be shared to CaCx clients as supported hereunder. Treatment Side Effects CaCx clients want to know information concerning the side effects of the different kinds of treatment lines they receive along their care journey. They noted that along treatment, both short and long-term side effects occur; something they need to be cognizant of since some effects turn out to lead to a deteriorating health. CaCx and its Causes. Many people seem not aware of this silent killer as it emerged that basic information relating to CaCx such as what it is, where it affects, who is at risk and the risk factors among others were not obviously known to them and needed to be shared to the public. The commonest mode of information sharing is by the word of mouth (face-face) and this does not facilitate adequate information sharing since it is affected by language barrier besides the clinic being a high-volume facility with small numbers of care providers. Conclusion Information needs slightly differed between CaCx clients and those of the general public yet between CaCx clients and healthcare professionals, they were comparable. Majority CaCx clients prefer to receive extensive information at their first diagnosis. Information about treatment, procedures as well as side effects, and prognosis are of particular relevance. Findings demonstrated that information seeking continues from diagnosis to follow-up. Therefore, healthcare providers listening to the clients might be more effective way to prevent and fight CaCx. It is also imperative to make use of the increasing adoption of ICTs to develop a centralized information portal on CaCx where all awareness information can easily be accessed by whoever might need it. Then efforts be made by referring and encouraging clients to use this portal as an effective way to further increase CaCx awareness.

6

Authors: Bonny E Balugaba, Simon K Ruttoh, Elizabeth K Ekirapa, Abraham A Siika, Martin C Were

Background and Purpose: While the benefits of electronic medical record systems (EMRS) are well known, the costs associated with their implementation in resource-constrained settings are not well documented. Establishing such costs is crucial for sound decision-making regarding allocation of scarce resources. This study aimed at conducting a comprehensive cost analysis for developing, implementing and maintaining an EMRS in a resource-constrained setting. Methods: The study was conducted at an urban HIV/AIDS clinic in Kampala, Uganda using an organizational perspective. A cost inventory was used to collect costs from original receipts and invoices, and vendor market prices used where receipts could not be accessed for the 8 years under study. All costs were converted to the equivalent US dollar value and those with a lifetime more than one year annuitized. A paper-based and an open-source record system were the comparators. Results: The total direct cost of the system was USD 1,066,965 while an open-source instance would cost just over USD 800,000. The paper-based system cost was half the cost of the EMRS (USD 544,159). Salaries and wages contributed 70% (USD 757,332) and 81% (USD 438,842) of costs in the EMRS and paper-based system, respectively. The other major cost driver was software and licensing (20%, USD 258,341). Conclusions: EMRS have higher costs as compared to paper-based medical record systems. However, expenditure on the system reduces with time. Salaries and wages are the largest contributors to the cost burden. Open-source systems have a potential to lower costs.

7

Authors: Jamiru Mpiima, Josephine Nabukenya

Background To effectively plan for human resources for health, health managers need to keep track of the size and composition of their human resources and to anticipate their future needs. Effective strategic planning is however informed by quality and reliable workforce information. However, in Uganda, existence and access to adequate information on human resources for health still remains a challenge. This limits the capacity of health managers to effectively monitor and make policies. To this end, this study set out to examine the challenges that affect the management of health human resources information in Uganda; with an aim to recommend a single authoritative source of health workforce information that can provide an accurate count of all health work-force in Uganda’s health system. Methods We used qualitative methods to examine the challenges that are affecting the management of health workers information in Uganda. Key informant interviews were conducted with purposively selected health-related organizations with a stake in human resources for health information management. Results The study reports various systems used for health workforce information management, the common one being the integrated Human Resources Information System (iHRIS). The study also provides various challenges affecting these systems including fragmentation and uncoordinated human-resource systems, which results to duplication of health workers records, incomplete or missing data elements, and inaccurate health workforce accountability. Conclusion Given the nature of the challenges, the study denotes the need for a single authoritative source of health workforce information that can provide an accurate count of all health workforces in Uganda’s health system commonly known as a National Health Worker Registry. The registry can be used to aggregate the key attributes for all health workers in the country. Thus our future study will focus on defining country-specific features and requirements that NHWR should poses.

9

Authors: Alpheus Wanano Mogwe

Background and Purpose: The aim of the research was to investigate and develop an e-health mobile application with associated algorithms for use in managing patient information collection. The research explored the current role of Information Technologies (ICTs) in the Botswana Public Healthcare System (BPHS). Specifically, it dealt with elucidating, discussing and understanding the current status of BPHS and ICTs role in BPHS in relation to the utilisation of Electronic Health Records (EHRs). The author explored the current BPHS ways of gathering and processing patient information, categorization of the current BPHS and users’ satisfaction levels with it and ascertained people’s behaviours towards employment and utilisation of ICTs in health. Methods: Triangulation of quantitative and qualitative methods was followed, with questionnaires, interviews and observation techniques employed to draw conclusions. Results: There was strong evidence that many users showed dissatisfaction with BPHS, which is mostly paper oriented and manual, and indicated an embracement of mobile applications in health. Conclusions: The study concluded that the current BPHS is mostly paper based and the satisfaction levels of users negate such systems in this modern era of utilisation of ICTs. In conclusion, the author designed and developed a working prototype for BPHS.

10

Authors: Catherine Kabahuma, Josephine Nabukenya

Background and Purpose: Managing accuracy to identify any patient is very crucial in ensuring continuity, quality and coordination of their healthcare over time. Patient identification management (PIM) in healthcare is still a challenge in Uganda for both public and private healthcare institutions including insurance companies using different mechanisms to uniquely identify clients. This deters PIM and individual health data linkage across healthcare organizations. An integral part of individual health data linkage and health information exchange (HIE) process across healthcare organizations is verification and/or validation of a client’s identity. This vital service can be provided by an administrative electronic health information system known as a client registry (CR). A CR is a central electronic database that holds patient identifying information and demographics with a mechanism to uniquely identify each of them using select identifying information such as unique identifiers within a given territory. Consequently, in preparation for the establishment of a National CR (NCR), we needed to first examine the current mechanisms used and the challenges faced in PIM in Uganda’s health facilities. The NCR is envisaged to facilitate and harmonize PIM and HIE in Uganda's healthcare system. Methods: A cross-sectional study was conducted to document the mechanisms used for and challenges facing PIM in Uganda and define NCR requirements. The Design Science Research (DSR) approach was employed to achieve the overall study aim with a much broader goal beyond this paper. Specific to this paper, the research question for which the methods and results presented was formulated under stage two of the DSR six-step process model. 35 key informant interviews with purposively selected representatives from select organizations and health facilities were conducted. Results: Majority of health facilities use either paper-based or a combination of both paper-based and electronic tools to register clients, manage their identities and data. Key challenges reported facing PIM relate to data storage, retrieval, client identity verification, and tracking. These translated into four key NCR requirements that can facilitate PIM and HIE in Uganda’s healthcare system. Conclusions: The study proves that there is need to address the various PIM challenges; as such our next steps are to establish actual NCR requirements/services in order to harmonize PIM across Uganda’s health facilities.

10

Authors: Chipo Kanjo, Joshua Hara, Jens Kaasbøll

Background and Purpose: Utilization of Electronic Medical Records (EMRs) bring many benefits to patient record handling. Previous research has shown that medical doctors in hospitals in Africa had poor IT skills, partly disabling their use of EMRs. Working in health centres, we wanted to know how to improve the digital skills of their personnel. Methods: Qualitative methods including interviews, informal discussions and observation were used for data collection. The study was conducted in three health facilities in Malawi, selected purposively as these were the ones implementing the EMR under study. The study targeted all staff that were capturing data at point of care using the app between January and June 2019. Results: Personnel in the health centres were computer illiterate, even if they used phones, and some smart phones. After computer training, they were still not comfortable using EMRs because their typing speed was delaying the process. Improving the typing speed allowed the users to comfortably use the EMRs. Three levels of EMR empowerment were identified; being able to 1) use EMR where typing speed is less relevant, 2) use EMR at point of care, 3) guide colleagues on their EMR use and find new ways of exploiting the technology. For reaching level 2, 50-100 hours of typing practice would be necessary. Conclusions: Health workers who lack computer skills tend to shy aware from using EMRs. Even if falling prices of equipment will enable more EMRs in rural areas, the digital empowerment process will still require significant resources.