This is an open-access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Public Health and Surveillance, is properly cited. The complete bibliographic information, a link to the original publication on http://publichealth.jmir.org, as well as this copyright and license information must be included.
In moving toward malaria elimination, one strategy is to implement an active surveillance system for effective case management. Thailand has developed and implemented the electronic Malaria Information System (eMIS) capturing individualized electronic records of suspected or confirmed malaria cases.
The main purpose of this study was to determine how well the eMIS improves the quality of Thailand’s malaria surveillance system. In particular, the focus of the study was to evaluate the effectiveness of the eMIS in terms of the system users’ perception and the system outcomes (ie, quality of data) regarding the management of malaria patients.
A mixed-methods technique was used with the framework based on system effectiveness attributes: data quality, timeliness, simplicity, acceptability, flexibility, stability, and usefulness. Three methods were utilized: data records review, survey of system users, and in-depth interviews with key stakeholders. From the two highest endemic provinces, paper forms matching electronic records of 4455 noninfected and 784 malaria-infected cases were reviewed. Web-based anonymous questionnaires were distributed to all 129 eMIS data entry staff throughout Thailand, and semistructured interviews were conducted with 12 management-level officers.
The eMIS is well accepted by system users at both management and operational levels. The data quality has enabled malaria personnel to perform more effective prevention and control activities. There is evidence of practices resulting in inconsistencies and logical errors in data reporting. Critical data elements were mostly completed, except for a few related to certain dates and area classifications. Timeliness in reporting a case to the system was acceptable with a delay of 3-4 days. The evaluation of quantitative and qualitative data confirmed that the eMIS has high levels of simplicity, acceptability, stability, and flexibility.
Overall, the system implemented has achieved its objective. The results of the study suggested that the eMIS helps improve the quality of Thailand’s malaria surveillance system. As the national malaria surveillance system, the eMIS’s functionalities have provided the malaria staff working at the point of care with close-to-real-time case management data quality, covering case detection, case investigation, drug compliance, and follow-up visits. Such features has led to an improvement in the quality of the malaria control program; the government officials now have quicker access to both individual and aggregated data to promptly react to possible outbreak. The eMIS thus plays one of the key roles in moving toward the national goal of malaria elimination by the next decade.
Malaria transmission occurs in all six World Health Organization regions. An estimated 3.3 billion people are at risk of malaria infection worldwide, and 1.2 billion are at high risk (>1 in 1000 chance of contracting malaria in a year). In 2013, there were 33,302 confirmed cases in Thailand, with 37 confirmed deaths [
The eMIS was originally developed by the Center of Excellence for Biomedical and Public Health Informatics (BIOPHICS) and Mahidol University via support from the World Health Organization and the Bill and Melinda Gates Foundation. The specific aim was to replace the traditional paper-based malaria reporting system with an active and close-to-real-time electronic reporting system [
The eMIS is operated and overseen by the Bureau of Vector Borne Disease (BVBD), which falls under the Department of Disease Control within the Thai Ministry of Public Health. This occurs through a network of malaria clinics and malaria posts located in villages in malaria-endemic areas. Between the highest decision-making level at the ministry and the local operating sites, there are regional offices of disease prevention and control, vector-borne disease control center (VBDC), and vector-borne disease control unit (VBDU). Currently, eMIS data are hosted at a secure server located at BIOPHICS, only accessible by authorized system users at the Ministry of Public Health. As the eMIS operates under local BVBD staff, BIOPHICS acts as a system developer and active technical system support for the ministry.
The eMIS platform uses offline and online replication technology to enable malaria staff to continue entering data offline when facing an unstable Internet connection. Whenever Internet is available, data can be later synchronized between data entry sites and the central database. Thus, there are some limitations regarding available computer hardware and Internet signal in certain remote areas and/or difficulties in performing simultaneously case management and online data entry. Therefore, each malaria record is captured through a hybrid process in which data are initially collected on paper and subsequently entered into the eMIS via a desktop offline/online client software platform (
Since its implementation, the eMIS has been evolving; the system has demonstrated its ability to capture essential data from individual malaria cases at local operational units and the data are being used for effective analysis and decision support at upper management levels [
Paper-based forms and electronic data entry screens.
Data flow of the electronic Malaria Information System (eMIS).
This is a descriptive study using a mixed-methods evaluation, combining quantitative and qualitative approaches. A number of previous studies have evaluated system innovations using mixed methods and confirmed that multiple evaluation methods can comprehensively identify system and/or electronic record usability, challenges, and specific problems [
Tak and Trat, the two border provinces closest to Myanmar and Cambodia, respectively, were purposively selected as provinces with the highest number of malaria cases along Thailand’s malaria-endemic borders. Source data in the paper-based forms (pCDF and pCIF) versus the electronic data records in the eMIS (eCDF and eCIF) were reviewed and compared. Overall, 4455 pCDFs and 784 pCIFs from the period December 2013 to January 2014 were collected from record books for the two provinces. Similarly, eCDF and eCIF for the same period were extracted from eMIS databases. A Web-based anonymous electronic questionnaire was distributed to all eMIS data entry staff at VBDC/VBDU; of all 129 data entry staff, 128 (99.22%) completed the questionnaire. Semistructured interviews were conducted with management-level users: 1 malaria surveillance officer at the ministry, 12 monitoring and evaluation officers (M&Es) at VBDC/VBDU in Tak and Trat provinces, and 1 information technology (IT) officer at BIOPHICS.
There are several methods available for the evaluation of the effectiveness of health care systems, innovation, or tools used for capturing electronic data records [
Attribute definitions and corresponding data collection approaches.
Attributes | Descriptiona | Data collection approaches |
Data quality | Completeness (absence of missing values) and validity (absence of errors) | Data record review and semistructured interview |
Timeliness | Delay in reporting | Data review |
Simplicity | Method of collecting data and time needed to collect data; structure of the system | Structured questionnaire |
Acceptability | Willingness of users to use the eMISb | Structured questionnaire |
Flexibility | Capacity to cope with new requirements and standards the system follows | Structured questionnaire and semistructured interview |
Stability | Downtime of server and response of technical support | Structured questionnaire and semistructured interview |
Usefulness | Dissemination of knowledge | Structured questionnaire and semistructured interview |
aBased on Centers for Disease Control and Prevention
beMIS: electronic Malaria Information System.
Data quality was assessed in terms of the number of records captured using both paper and electronic data collection methods, data completeness or missing data, data validity or consistency between the data in paper and electronic formats, and logical errors or conflicting values among data elements. Timeliness was assessed in terms of delays in data entry from paper data collection forms into electronic records in the eMIS. Both data quality and timeliness were evaluated by comparing the electronic data records in the eMIS (eCDF and eCIF) against the paper data collection forms (pCDF and pCIF). The data from paper-based forms were entered using Microsoft Excel 2013 and double-checked for comparisons with electronic records in the eMIS.
Other surveillance attributes were assessed via a structured questionnaire and semistructured interviews. To fit within the eMIS context, the questionnaire was adapted from a previous study that measured similar attributes concerning data quality in a surveillance system [
This study was approved by the Ethics Committee of the Faculty of Tropical Medicine, Mahidol University. This study did not involve vulnerable participants. The participants were informed and provided their consent after reading documents explaining the purpose of the study, the participant’s risks and benefits, and the confidentiality and protection of their data. The participants were informed that their answers to the questionnaire were anonymous and would not affect their employment.
In general, based on qualitative data collection methods, the data quality of the eMIS was perceived as “good” among management-level users. Officers at the BVBD stated that the data quality of the eMIS was acceptable and better than paper reports, despite no previous evaluation. All M&Es stated that they check the data entered in the eMIS against the paper forms; however, they only check the aggregated case number. One M&E mentioned that the eMIS data were complete while others claimed the data were more accurate than the paper forms.
I think the data quality is acceptable but needs to be improved. But it’s better than paper reports.
The eMIS helps me a lot. The data are more accurate than paper.
In evaluating data quality based on the number of data records being reviewed, the total number of pCDFs did not match eCDF numbers: 38,860 noninfected cases were reported using 4455 pCDFs (several cases were listed on a single paper page) but 41,451 records (one case is one record) were found as eCDFs in the eMIS database. Similarly, fewer infected cases were found in the paper records (one case per single paper page) than those recorded electronically (one case is one record): 781 pCDFs versus 964 eCDFs and 784 pCIFs versus 969 eCIFs.
Completeness of core data elements (data fields) in the CDFs and CIFs was only checked among infected cases as shown in
To evaluate data validity regarding the consistency of the core data elements in the paper and electronic forms, a linkage between the two data entry forms was made. Adapting the data matching method suggested in the literature concerning the assessment of data quality in a cancer registry [
Completeness of core elements.
Core data elements | Total missing | % missing | 95% CI | Total missing | % missing | 95% CI |
pCDFa(n=781 records) | eCDFa(n=964 records) | |||||
Date of form entry | - | - | - | 0 | 0 | - |
Date of drawing blood | 8 | 1.02 | 0.32-1.73 | 0 | 0 | - |
Date of receiving blood film | 186 | 23.82 | 20.83-26.80 | 0 | 0 | - |
Date of blood test | 173 | 22.15 | 19.24-25.06 | 0 | 0 | - |
Type of blood test | 6 | 0.77 | 0.16-1.38 | 0 | 0 | - |
Type of patient | 4 | 0.51 | 0.01-1.01 | 0 | 0 | - |
Area classification | 266 | 34.06 | 30.74-37.38 | 0 | 0 | - |
Nationality | 17 | 2.18 | 1.15-3.20 | 0 | 0 | - |
Patient name | 0 | 0 | - | 0 | 0 | - |
Age | 0 | 0 | - | 0 | 0 | - |
Sex | 5 | 0.64 | 0.08-1.20 | 0 | 0 | - |
Blood test result | 0 | 0 | - | 0 | 0 | - |
Medicine | 6 | 0.77 | 0.16-1.38 | 0 | 0 | - |
pCIFa(n=784 records) | eCIFa(n=969 records) | |||||
Date of form entry | - | - | - | 0 | 0 | - |
Date of drawing blood | 1 | 0.13 | 0-0.38 | 0 | 0 | - |
Date of blood test | 1 | 0.13 | 0-0.38 | 0 | 0 | - |
Date of investigation | 1 | 0.13 | 0-0.38 | 0 | 0 | - |
Area classification | 142 | 18.11 | 15.42-20.81 | 0 | 0 | - |
Nationality | 22 | 2.81 | 1.65-3.96 | 0 | 0 | - |
Patient name | 0 | 0 | - | 0 | 0 | - |
Age | 3 | 0.38 | 0-0.81 | 0 | 0 | - |
Sex | 7 | 0.89 | 0.23-1.55 | 0 | 0 | - |
Blood test result | 1 | 0.13 | 0-0.38 | 0 | 0 | - |
Case classification | 53 | 6.76 | 5.00-8.52 | 0 | 0 | - |
Infection location | 65 | 8.29 | 6.36-10.22 | 0 | 0 | - |
apCDF: paper case detection form; eCDF: electronic case detection form; pCIF: paper case investigation form; eCIF: electronic case investigation form.
Agreement between paper-based and electronic data.
Core data elements | No. of discordant pairs | Percentage of disagreement | 95% CI |
Date of drawing blood | 12 | 1.69 | 0.74-2.63 |
Date of receiving blood film | 183 | 25.74 | 22.52-28.95 |
Date of blood test | 180 | 25.32 | 22.12-28.51 |
Type of blood test (ACD/PCD) | 8 | 1.13 | 0.35-1.90 |
Type of patient (new case/follow-up) | 3 | 0.42 | 0-0.9 |
Area classification (A1, A2, B1, B2) | 249b | 51.88 | 47.41-56.34 |
Nationality (Thai, M1, M2) | 21 | 2.95 | 1.71-4.20 |
Age | 7 | 0.98 | 0.26-1.71 |
Sex | 7 | 0.98 | 0.26-1.71 |
Blood test result (type of malaria) | 2 | 0.28 | 0-0.67 |
Medicine | 7 | 0.98 | 0.26-1.71 |
Area classification (A1, A2, B1, B2) | 260 | 36.16 | 32.65-39.67 |
Nationality (Thai, M1, M2) | 28 | 3.89 | 2.48-5.31 |
Age | 26 | 3.62 | 2.25-4.98 |
Sex | 16 | 2.23 | 1.15-3.30 |
Date of drawing blood | 10 | 1.39 | 0.53-2.25 |
Date of blood test | 9 | 1.25 | 0.44-2.06 |
Blood test result (type of malaria) | 13 | 1.81 | 0.83-2.78 |
Date of investigation | 60 | 8.34 | 6.32-10.37 |
Case classification | 66 | 9.18 | 7.07-11.29 |
Infection location (within village/cottage/forest) | 118 | 16.41 | 13.7-19.12 |
aACD: active case detection; pCDF: paper case detection form; eCDF: electronic case detection form; pCIF: paper case investigation form; eCIF: electronic case investigation form; PCD: passive case detection; M1: migrant group 1; M2: migrant group 2.
bDenominator varied because some paper-based blood record forms did not include this item.
A further factor to measure data quality was evaluated: logical errors in the data elements. Logical errors were assessed in two ways: (1) illogical chronological sequence of data capture and reporting dates and (2) discrepancies in reporting malaria prescriptions against national standard guidelines for each type of malaria infection. It should be noted, however, that there were a number of incomplete dates on pCDFs because some infected cases were detected at district or provincial hospitals rather than at the malaria clinics or posts, and hospital health care personnel may not have provided detailed data concerning detection dates (eg, date of receiving blood film, date of blood test) for those infected cases. Thus, complete information regarding the dates of those cases on pCDFs was based on just 593 cases. However, the CIFs were completed and all 784 pCIFs and 969 eCIFs for infected cases were used in the analysis. First, discrepancies in chronological order were checked for different time points (
Another illogical error examined in this study concerned discrepancies in medication prescriptions (type of medicine provided). In Thailand, malaria treatment is regulated and only available at governmental health care facilities for both Thai and non-Thai patients across malaria-endemic areas. Specific standard regimens for each type of malaria infection, mainly
Error rate of date sequence inconsistencies in case detection forms (CDFs) and case investigation forms (CIFs).
In exploring timeliness, the descriptive statistics shown in
Some papers arrive late, so we report late.
Timeliness analysis.
Reporting delay (days) | Mean (SD) | Median (minimum-maximum) |
Delays between date of blood drawn and date of form entry for eCDFa(n=963 recordsb) | 9.34 (10.64) | 4 (0-69) |
Delays between date of investigation and date of form entry for eCIFa(n=906 recordsc) | 9.14 (16.27) | 3 (0-134) |
aeCDF: electronic case detection form; eCIF: electronic case investigation form.
bOne record with more than a 200-day delay (242 days) was considered an outlier and excluded.
cFive records with more than 200 days (369, 372, 372, 366, and 242 days) were considered as outliers and excluded; 58 records with negative delays were excluded, ranging from −1 to −15 days.
The simplicity of the eMIS refers to both its structure and ease of operation. The eMIS represents a single, vertical, streamlined data entry process. It has a simple and vertical information structure with one central data center at BIOPHICS. All data entry staff report that the time required to enter eMIS data was easy to incorporate into their daily work duties. A total of 85/128, 66.40% of respondents reported spending less than 5 minutes entering into the eCIF the details of 1 infected case (with approximately 80 data elements), and much less time completing 1 CDF via eCDF (42 data elements for a noninfected case and 52 for an infected case). Respondents reported that the time burden for data entry was manageable.
The majority agreed or strongly agreed that the eMIS is easy to use (119/128, 92.97%), the report forms in the eMIS are easy to complete (112/128, 87.5%), instructions for completing the forms are clear and helpful (113/128, 88.28%), help is easy to access (114/128, 89.06%), and that learning to operate the program is easy (116/128, 90.63%). A slightly lower percentage of respondents agreed or strongly agreed that the eMIS has a user-friendly interface (83/128, 64.84%), with 31.25% (40/128) feeling neutral about this statement (
The data entry part of the system should be labeled with colors, especially in fields which have been entered.
The screen for CIF data entry should be separated according to each part of the CIF form. It could use lines or colors to separate each screen for more convenient data entry.
Simplicity of the electronic Malaria Information System (eMIS).
Acceptability refers to the willingness of users and organizations to use the system on a daily basis. Initially, the eMIS was implemented in 7 provinces along the Thai-Cambodia border but now includes 38 provinces in Thailand. As of mid-2015, access to the eMIS is expanding to include a number of subdistrict health centers in malaria-endemic provinces.
A significant majority of eMIS data entry staff agreed or strongly agreed that their contribution to the eMIS is valuable (127/128, 99.22%), they enjoy being involved with the malaria control progress via the eMIS (126/128, 98.44%), and that their contribution to the reporting of malaria cases to eMIS was adequately acknowledged (120/128, 93.75%). Most (122/128, 95.31%) agreed that the eMIS is very important. Most data entry staff (121/128, 94.53%) felt that they received support from the eMIS support team (
Acceptability of the electronic Malaria Information System (eMIS).
Flexibility reflects how the system copes with changes. Initially, the eMIS was implemented in 7 provinces as an online system; users were only able to enter data when connected to the Internet. However, because of an unstable Internet connection, the eMIS was changed to include an offline mode. After this big change, eMIS underwent three other main revisions.
We always maintain the core structure and do minor changes according to users’ requirements. Technically, eMIS adapts to change very well.
However, the eMIS is not based on any health information standard, such as SNOMED CT, ICD-10, or Health Level Seven. An IT manager mentioned that it would not be difficult to apply some standard code to ensure better data integration in the future.
eMIS only focuses on one disease, malaria, so we didn’t apply any health information standard. But we think it would be good if we applies a standard code in order to integrate our data with the national disease report system in the future.
Stability reflects the reliability and availability of the eMIS. In all, 100/128, 78.13% of data entry staff reported experiencing some technical problems with the eMIS. The most frequently reported technical problem was data synchronizing, with 87/128, 67.97% of respondents stating they had difficulty synchronizing offline data with the eMIS. This is consistent with interviewed respondents, with 6 M&Es (50%) mentioning the issue of data synchronizing. One M&E specified that data synchronizing was slow, especially at the end of month when many users attempt to upload data at the same time. When asked about technical support for the eMIS, 116/128, 90.63% agreed or strongly agreed the technical issues were dealt with quickly and efficiently.
In terms of reliable resources, the eMIS currently receives funding from the Global Fund and is seeking long-term governmental funding.
At the beginning, eMIS was supported by the Global Fund. Now we plan to use domestic funds. We have both the Global Fund and government funds for staff salaries. In the future, all salaries will be covered by domestic funds.
We have a limited budget for IT after the Global Fund stopped its support. We need a budget for new computers.
Now the eMIS is very good. But we hope the eMIS will be a long-term program. We need funding to support our IT staff after the Global Fund stops its support.
The eMIS platform appears stable and performs well with no records of system failure. Data stored in the eMIS is backed up daily to tape media.
The eMIS has never been down since implementation. We have installed and configured virtual machine technology that can perform live migration for the eMIS. Live migration can move a running virtual machine (web and database server) from one physical server to another without any effect to the users.
By the end of 2016, it is hoped that the management of the eMIS functional hardware and software architecture will be transferred from BIOPHICS to the BVBD at the Ministry of Public Health. One M&E expressed some concern over this change.
In the future, the management of the eMIS will be led by BVBD. My concern is who will take care of the IT problems. Now BIOPHICS do a great job of IT support. And I worry about the salary of IT staff that we will hire for the eMIS in the future.
The eMIS is considered a powerful tool and is used by the BVBD in its aim toward eliminating malaria in Thailand. With data from the eMIS, the BVBD is able to change policies, allocate funds, and prioritize malaria elimination activities. The BVBD provides weekly reports and presentations based on eMIS data. An officer at the BVBD reported that the eMIS can influence and guide future research.
We use the results from eMIS to plan elimination. It has helped us to change policy. Before we didn’t have data to indicate how to implement malaria elimination. Now we are involved in subregional elimination. The data from the eMIS help us to target where we are going to eliminate malaria and apply a budget, and also to verify where malaria has been eliminated. eMIS is a powerful tool, without it, I don’t think we can ensure malaria elimination. It’s a key tool for malaria elimination in Thailand to change policy.
At a lower level, all M&Es described the eMIS as being very useful, and considered the eMIS as integral to their routine work. It also provides useful information on malaria patients and tracks malaria patients to ensure effective action. The most commonly reported functions were data analysis, data presentation, and trend analysis. With the eMIS, M&Es are able to track patients, efficiently manage and control any outbreaks, perform campaign and training tasks, and make future plans.
For new cases, we take action quickly and we can predict the management for the next month. With the eMIS, we can see the malaria cases easily and make plans for the future.
eMIS helps to track patients. We can see the nearby situation and we can get prepared. We can see the real-time situation. We are able to plan how to control before an outbreak and understand the endemic. We can do an analysis for control and prevention. eMIS can quickly transfer data to the VBDU according to the patient’s address. eMIS provides information on the outbreak of disease and accurate patient addresses.
We compare the malaria cases between different years to see trends in malaria outbreaks. We can track cases and control these cases with well-timed action.
After its development and implementation throughout malaria-endemic areas in Thailand, the eMIS has fully become the national malaria surveillance system. The eMIS appears to be well accepted by system users at both management and operational levels. The reason for its success could be that the design of the eMIS is not overly complex; it still combines original paper-based data collection forms with electronic records. The system flow is practical and can be used as a standard for health care activities in limited-resource settings. Some studies have shown that although paper forms can serve as an important tool for health care personnel in their work, their use can circumvent the intended system design. The following categories describe the benefits of using a paper-based system despite an electronic information system designed for capturing as electronic records: (1) efficiency; (2) knowledge, skill, ease of use; (3) memory; (4) sensorimotor preferences; (5) awareness; (6) task specificity; (7) task complexity; (8) data organization; (9) longitudinal data processes; (10) trust; and (11) security [
In general, data quality was acceptable. The quality of data has provided useful information for malaria personnel to ensure a high level of prevention and control activities. As mentioned above, there were some discrepancies regarding the number of source data forms (pCDF, pCIF) and electronic data records in the eMIS (eCDF, eCIF) for both noninfected and infected cases. There are a number of possible reasons for this result. First, the filing system for the source documents (pCDF, pCIF) was not well organized; some forms may have been misplaced or lost at the time of data collection. Second, as mentioned before, many noninfected cases were captured from the other sources (ie, district or provincial hospitals beyond BVBD primary care units); these were entered directly into the eMIS and thus there were no source documents (pCDF) for such cases. However, almost all infected cases have a source document (pCIF), no matter from which health care facility the data were obtained.
A small number of chronological errors were observed with the pCDF but not for eCDF; however, date errors were found for both pCIF and eCIF. In general, fewer chronological errors were seen in electronic records than on paper-based forms. This type of error might occur because of human error during the filling out of the paper source document and/or data entry into the system. Where the date of the eCIF entry was earlier than the date of investigation, this could be explained as a transcription error in the data entry process. This finding concurs with that suggested in the literature—should be the use of personal digital assistant based data collection or entry system could increase efficiency and reduce data transcription errors for public surveillance data collection in developing countries [
Another logical check conducted in this study revealed that there were incorrect prescriptions of malaria treatment in both paper-based and electronic data. However, this result should be interpreted with care. This does not mean that the patients received the inappropriate medication; it simply reflects that some patients received treatment that differed from the national standard guidelines. In the eMIS, data elements regarding medication prescriptions are more extensive than the treatment list provided on paper forms. Thus, the conflicting statistics found in this study could occur because some hospital reports did not apply the same malaria treatment as that in the national guidelines.
In this study, a high level of completeness was identified for electronic data records and is consistent with other studies [
The timeliness analysis indicated that the median time from data captured in paper forms at point-of-care units to data entry as electronic records at VBDC/VBDU was approximately 3-4 days. A previous study on the timeliness of a public health surveillance system [
Respondent eMIS users indicated that the system is simple, with the eMIS based on the routine workflow of a vertically structured health care system in Thailand. Most users (both management and operational) accepted the eMIS and were satisfied with its performance; they considered the system easy to incorporate into other work duties. Furthermore, their commitment to be part of the eMIS was high. The M&Es reported being satisfied with the data analysis function and BVBD valued the eMIS as having a key role in the malaria elimination process. Factors influencing the effectiveness of the system have been frequently mentioned in the literature and include trust in quality of the data, motivation of the system users, and outcome expectancy [
In assessing the usability of the system, one approach is to define usability as how useful, usable, and satisfying a system is for intended users [
Infrastructure is the backbone of every system, and the successful adoption of any eHealth system depends on the infrastructural arrangements [
Overall, this evaluation, based on data from two malaria-endemic provinces for the period December 2013 to January 2014, has confirmed that eMIS is achieving its objective as an effective platform. The data quality assessment via an intensive look at the data records in the system flow, and the conduct of a Web-based questionnaire survey with all data entry staff throughout Thailand and the in-depth interview with data users from lower level to management level, have suggested that the eMIS helps improve the quality of Thailand’s malaria surveillance system. As the national malaria surveillance system, the eMIS’s functionalities have provided the malaria staff working at the point of care with close-to-real-time case management data quality, covering case detection, case investigation, drug compliance, and follow-up visits. Having access to such information in forms of individual case report, aggregated data in the geographical mapping, and data visualization platform could lead them to respond in a timely manner to the situation in their areas of responsibility. The benefits of eMIS, particularly in assessing the malaria situation, are recognized as exceeding those of the original paper-based reporting system alone. In other words, the eMIS is an information system that supports Thailand’s malaria surveillance. One of the system features that makes the system users satisfied was that it can bring more evidence in the individual practice, for instance by providing electronic reminders about actions to take or treatments to prescribe at point of care (both triggered by previously keyed information). It has also led to an improvement in the quality of the malaria control program because case management is now somewhat standardized and government officials can have quicker access to both individual and aggregated data to promptly react to a possible outbreak. With such features of the system, the eMIS plays one of the key roles in moving toward the national goal of malaria elimination by the next decade.
Center of Excellence for Biomedical and Public Health Informatics
Bureau of Vector Borne Disease
electronic case detection form
electronic case investigation form
electronic Malaria Information System
information technology
monitoring and evaluation officer
paper case detection form
paper case investigation form
vector-borne disease control center
vector-borne disease control unit
vector-borne disease control office
This project was funded by a grant from The Rockefeller Foundation and the Faculty of Tropical Medicine, Mahidol University. The authors would like to gratefully extend thanks to the funders for supporting the graduate students in the biomedical and health informatics program at Mahidol University. The authors also wish to thank the eMIS and data management teams at BIOPHICS and staff at Thai Ministry of Public Health for their contributions to this study.
SM and JK designed and planned the study, drafted the first version of the paper, submitted the paper, and approved the final version. SL, NS, and BK assisted in designing and planning the study, wrote the submitted paper, and approved the final version. SM, JK, KJ, and AK managed the literature search as well as performed the statistical analysis. RP, SJ, KT, AK, and KJ supervised data collection at the study sites. All authors read and approved the final manuscript.
None declared.