District Health Information Software, or DHIS2, is one of the most important sources of health data in low- and middle-income countries (LMICs). Used by 72 different LMIC governments, DHIS2 is a web-based open-source platform that is used to collect, manage, and analyze routine and critical health data. It is the backbone for national health information systems in these countries and a vital resource for monitoring program and policy performance.

區衛生信息軟件，或DHIS2 ，是在低收入和中等收入國家(低收入國家)的健康數據的最重要來源之一。 DHIS2由72個不同的LMIC政府使用，是一個基于Web的開源平臺，用于收集，管理和分析常規和重要的健康數據。 它是這些國家國家衛生信息系統的骨干，也是監控計劃和政策績效的重要資源。

Since developing DHIS2, the Health Information Systems Program at the University of Oslo has worked tirelessly to improve the platform to respond to user needs and support training in the platform. The global health community has simultaneously invested in initiatives to improve DHIS2 data quality and encourage the use of DHIS2. However, there has been less attention towards improving the capacity to visualize data within DHIS2. I’ve previously described how often in global health, dashboards can be treated as go-to solutions without essential reflection on user interpretability and key messages; DHIS2 reflects this same quandary.

自開發DHIS2以來，奧斯陸大學的健康信息系統計劃一直在不懈地努力，以改進該平臺以響應用戶需求并支持該平臺中的培訓。 全球衛生界同時投資于改善DHIS2數據質量和鼓勵使用DHIS2的計劃。 但是，很少有人關注提高DHIS2中的數據可視化能力。 前面我已經描述了在全球健康狀況中， 儀表板可以被視為首選解決方案，而不必對用戶的可解釋性和關鍵信息進行實質性的思考。 DHIS2反映了同樣的難題。

Given the widespread use of DHIS2, it has become a key resource for COVID-19 surveillance in LMICs. The recent Digital Solutions for COVID-19 report by the Johns Hopkins University’s Global mHealth Initiative identified DHIS2 as one of the two most standout platforms for COVID-19 surveillance based on maturity, flexibility, and large-scale deployment. DHIS2’s virtually irreplaceable role in COVID-19 surveillance highlights not just the importance, but also the urgency, of considering how to improve data visualization within the platform. In 2018, Aprisa Chrysantina and Johan Ivar S?b? from University of Oslo conducted a study to assess the quality of user-created DHIS2 dashboards in Indonesia. The team has kindly shared some insights from their study with me below as they make a case for investing in data visualization within global health.

鑒于DHIS2的廣泛使用， 它已成為LMIC中COVID-19監視的關鍵資源 。 約翰·霍普金斯大學(Johns Hopkins University)的全球移動醫療計劃 ( Global mHealth Initiative)最近發布的COVID-19數字解決方案報告指出，基于成熟度，靈活性和大規模部署，DHIS2是用于COVID-19監視的兩個最出色的平臺之一。 DHIS2在COVID-19監視中幾乎不可替代的作用不僅凸顯了考慮如何改善平臺內數據可視化的重要性，而且凸顯了其緊迫性。 2018年，奧斯陸大學的Aprisa Chrysantina和Johan IvarS?b?進行了一項研究，以評估印度尼西亞用戶創建的DHIS2儀表板的質量 。 小組在下面與我分享了他們的研究中的一些見識，為他們在全球衛生領域投資數據可視化提供了依據。

Tricia Aung: Why is data visualization important to DHIS2?

Tricia Aung：為什么數據可視化對DHIS2很重要？

Aprisa Chrysantina and Johan Ivar S?b? (DHIS2): Visualisation is a big part of the platform. The users can visualise and analyse their data using pivot tables, all kinds of charts from bar, pie, line, to speedometer, and also GIS functionality. The DHIS2 team believes that the end point of having data is to use it (albeit, correctly) and it will not be possible without ability to visualise the data.

Aprisa Chrysantina和Johan IvarS?b?(DHIS2)： 可視化是平臺的重要組成部分。 用戶可以使用數據透視表，從條形圖，餅形圖，折線圖到速度計的各種圖表以及GIS功能來可視化和分析其數據。 DHIS2團隊認為擁有數據的目的是使用它(盡管正確)，并且沒有可視化數據的能力是不可能的。

How has DHIS2 been used in Indonesia?

DHIS2在印度尼西亞如何使用？

Indonesia has been implementing DHIS2 to integrate health data from different programs such as HIV, TB, and reproductive, maternal, newborn, and child health. Dashboard implementation has been central to the implementation as it allows health staffs to directly visualise the data they have collected. Less than 2 years after the pilot in the country, DHIS2 has been implemented or at least introduced in 127 districts (24.7% of 514 districts) in Indonesia, and has expanded from 6 pilot programs to 17 different work streams, including pharmaceutical and medical devices, human resources, home care, and COVID-19. The application is mostly used to integrate, visualise, analyse, and report aggregate data across various levels, from facility to district to province, and even national levels.

印度尼西亞一直在實施DHIS2，以整合來自不同計劃(如HIV，TB和生殖，孕產婦，新生兒和兒童健康)的健康數據。 儀表板實施一直是實施的中心，因為它允許衛生人員直接可視化他們收集的數據。 在印度開展試點活動不到兩年的時間，DHIS2已在印度尼西亞的127個地區(514個地區的24.7％)實施或至少引入，并已從6個試點計劃擴展到17種不同的工作流程，包括制藥和醫療設備，人力資源，家庭護理和COVID-19。 該應用程序主要用于集成，可視化，分析和報告跨各個級別的匯總數據，從設施到地區再到省甚至國家層面。

What inspired your dashboard assessment in Indonesia?

是什么激發了您在印度尼西亞的儀表板評估？

We understood that appropriate and relevant dashboards are not straightforward to make. Users need a certain level of data literacy to create, read, and analyse charts. There was limited evidence in the literature regarding the quality of dashboards created by health staff in the field.

我們知道，適當而相關的儀表板并不容易制作。 用戶需要一定水平的數據知識才能創建，讀取和分析圖表。 關于該領域的衛生人員創建的儀表板質量的文獻證據有限。

What led to your decision to use Stephen Few’s dashboard design criteria in the study?

是什么導致您決定在研究中使用Stephen Few的儀表板設計標準的？

When we conducted this study, guidance to create dashboard and data visualisation are often not presented systematically. However, in his book Information Dashboard Design, Few lists common mistakes with dashboard creation, ranging from clutter and inappropriate contextualisation, to outright wrong use of visualisation techniques. We identified that these mistakes could be translated into guiding assessment questions.

當我們進行這項研究時，創建儀表板和數據可視化的指導通常沒有系統地提出。 但是，在他的《 信息儀表板設計》一書中，很少有人列出儀表板創建中的常見錯誤，從混亂和不適當的上下文環境到完全錯誤地使用可視化技術。 我們發現這些錯誤可以轉化為指導性評估問題。

What do you think are the most important findings? Was there anything that surprised you?

您認為最重要的發現是什么？ 有什么讓您感到驚訝的嗎？

Aprisa was (ironically) glad that her suspicion of dashboard quality was proven and that we now have evidence that support that. We were surprised that the frequency of the inappropriate data visualisation usage problem was higher than we expected to be. Unfortunately, although we were involved in some of the trainings where these users created the dashboards, we didn’t investigate the reasons nor the creation process specifically for the purpose of the research. Also, we were surprised that these dashboards and charts were created by health information systems (HIS) consultants or government HIS staff which had either Bachelors or Masters degrees in clinical health/public health or IT. This implied that these advanced educational backgrounds did not guarantee data visualisation literacy.

Aprisa(具有諷刺意味的是)很高興她對儀表板質量的懷疑得到了證明，并且我們現在有證據支持這一點。 我們感到驚訝的是，不適當的數據可視化使用問題的發生頻率比我們預期的要高。 不幸的是，盡管我們參與了這些用戶創建儀表板的一些培訓，但我們并未針對研究目的調查原因或創建過程。 同樣，令我們感到驚訝的是，這些儀表板和圖表是由具有臨床衛生/公共衛生或IT學士學位或碩士學位的健康信息系統(HIS)顧問或政府HIS人員創建的。 這意味著這些高級教育背景不能保證數據可視化素養。

Did you notice any variation in dashboard quality across levels of the health system (e.g. national vs. facility)?

您是否注意到衛生系統各個級別(例如國家與機構)的儀表板質量有何不同？

We didn’t specifically address this question in our study. However the people we mentioned above created dashboards and/or charts in different scopes, whether it was for facilities or at the national level. Through my field visits and discussions, I also found that many of these dashboards were developed as part of DHIS2 training process with trainer/facilitator(s) being around. But as we didn’t investigate the training process and trainer’s roles in data viz creation in depth, we could not say anything further about it.

在我們的研究中，我們沒有專門解決這個問題。 但是，我們上面提到的人員在不同的范圍內創建了儀表盤和/或圖表，無論是用于設施還是在國家一級。 通過我的實地訪問和討論，我還發現許多儀表板是作為DHIS2培訓過程的一部分開發的，培訓師/輔導員在身邊。 但是，由于我們沒有深入研究培訓過程和培訓者在數據即數據創建中的作用，因此我們無法對此進一步說明。

How have you used the study findings?

您如何使用研究結果？

As the research highlighted that people are making poor chart type selection, we have implemented new dimensions for selection panels in the Data Visualization app in DHIS2. Thus, we expect that it will be easier to choose the data and it will guide the chart selection.

由于研究突出表明人們在選擇圖表類型方面比較差，因此我們在DHIS2的“數據可視化”應用程序中為選擇面板實現了新的尺寸。 因此，我們希望選擇數據會更容易，它將指導圖表的選擇。

The research has become a foundation for us to encourage the use of standardized dashboards that are built around standard operating procedures and shared down to users. This way, users will get more practically useful dashboards.

該研究已成為我們鼓勵使用圍繞標準操作程序構建并共享給用戶的標準化儀表板的基礎。 這樣，用戶將獲得更加實用的儀表板。

The dashboard sharing function has been improved as well. We also added descriptions and dashboards items like boxes that will allow dashboard creators to input more instructions and guidance on how to use the dashboards directly on the dashboards.

儀表板共享功能也得到了改進。 我們還添加了描述和儀表板項目(例如框)，使儀表板創建者可以輸入更多有關如何直接在儀表板上使用儀表板的說明和指導。

Specifically in Indonesia, we developed data visualisation guidance materials based on this study result that use real-life examples (i.e of common problems). We delivered these materials academic lectures, roll out, and refresher trainings in discussion method.

特別是在印度尼西亞，我們根據該研究結果開發了數據可視化指導材料，這些材料使用了現實生活中的例子(即常見問題)。 我們提供了這些材料的學術講座，推廣和討論方法方面的進修培訓。

In addition, the WHO meta-data packages primarily focuses on “best practice” data analysis through dashboards. Working with the WHO, we also make accessible pre-defined dashboards based on current knowledge on how to manage various health programs. These may or may not be fully compatible with what countries are currently collecting, but we also see that these are used as inspiration to make better dashboards in countries.

此外，世衛組織元數據包主要側重于通過儀表板進行“最佳實踐”數據分析。 我們還與世界衛生組織合作，根據有關如何管理各種衛生計劃的最新知識，提供了可訪問的預定義??儀表板。 它們可能與國家目前正在收集的內容完全兼容，也可能不完全兼容，但是我們也看到這些被用作啟發，以在國家中制作更好的儀表板。

This interview has been lightly edited for clarity.

為了清楚起見，對這次采訪進行了少量編輯 。

Thank you to Aprisa Chrysantina and Johan Ivar S?b? for participating in this interview. You can learn more about DHIS2 here. Special thanks to Senthil Natarajan for his editorial support.

感謝 Aprisa Chrysantina 和 Johan IvarS?b? 參加了這次采訪。 您可以 在此處 了解有關DHIS2的更多信息 。 特別感謝 Senthil Natarajan 的編輯支持。

Tricia Aung is a Research Associate and Faculty member at Johns Hopkins School of Public Health in the Department of International Health. She leads workshops and research in visualizing global health data for decision-making in low- and middle-income country audiences. She is a Co-Chair for the DVS Diversity Committee.

特里西婭·昂 ( Tricia Aung) 是 國際衛生部 約翰霍普金斯大學公共衛生 學院的研究員和教授 。 她領導著研討會和研究工作，以可視化全球衛生數據為中低收入國家的受眾提供決策依據。 她是DVS多樣性委員會的聯席主席。