Skip to content

COVID-19 and health worker infections: The need for disaggregated intersectional data

During world health worker week, Oluwatobi Ogundele and Margaret Walton-Roberts take stock of what the pandemic has done to health workers.

This world health worker week marks one year of the global COVID-19 pandemic. Our ability to manage the medical complications of COVID-19 have rested on backs of health care workers, and it shows. Physicians, especially women, face burnout, nurses are traumatized by the experiences they have faced, and in the long term care sector, reflecting longer term structural investment deficits, care workers have been disproportionately affected. Prior to the pandemic there were already concerns about burnout, or, as it is more appropriate to call such occupational distress, moral injury—the impossibility of healthcare workers meeting their professional obligations within unsustainable, inequitable and unjust healthcare systems. On top of this, the pandemic has tested health systems globally.

In response to this reality, we have seen several campaigns to honour our #healthcareheroes. So, we are taking stock of what the pandemic has done to health workers during this world health worker week.

How well have we measured what affects those we treasure?

Well, the answer to this question is not straightforward, as this one example from the US reveals. In February 2020 anesthesiologist Claire Rezba set up the Twitter account ‘US HCWs lost to Covid19’ @CTZebra and started tracking healthcare workers lost to COVID-19, she said “it just felt personal”, that there was limited official response, and she wanted to bear witness. The account shares memorials of US healthcare workers who have died of COVID-19, and by March 2021 it had listed over 3,400 deaths. As of 23 March 2021 the CDC’s COVID Data tracker, on the other hand, listed 1,497 healthcare personnel deaths. This contrast between a Twitter account, and CDC’s data tracker reflects the problem of effectively monitoring pandemic outcomes in the absence of systemic and standardized surveillance systems.

This data quality issue is compounded by several factors. First, we have limited sex disaggregated data on COVID-19 infection and mortality numbers generally, and especially for healthcare workers. Second, racialized and ethnic minority health workers are disproportionality affected by COVID-19, but data does not always capture this.  Third, this is a global data problem, since of 192 countries providing data only 54% report some kind of sex disaggregated  data, never mind other relevant factors. These data weaknesses undermine our ability to understand the differential risks the global health workforce have faced. What we do know is that  over 70% of the health and social care workforce are women, and in many OECD nations significant numbers of health workers are immigrants. We must understand the risks and vulnerabilities health workers face so that we can improve conditions of work and better protect healthcare workers, and to do this we must use an intersectional lens.

Intersectionality and health vulnerabilities

Intersectionality highlights how inequality is shaped by multiple intersecting factors including age, sex, gender, health status, geographic location, disability, migration status, race/ethnicity, and socioeconomic status, and how these can be multiplied and exacerbated in contexts marked by systematic inequalities. It is evident that COVID-19 infection and mortality disproportionally affects racialized communities and exacerbates already present inequities, and this is evident for healthcare workers. For example, in the UK and USA Black, Asian, and minority ethnic healthcare workers had an increased risk of COVID-19 compared with non-Hispanic white healthcare workers, and were the category of workers who reported having to reuse PPEs because of an inadequate supply. In the UK, as of April 2020, 71% of nurses and midwives who had died from COVID-19 were Black, Asian, and Minority Ethnic. Data from the UK show that more Filipino health care workers died of COVID-19 in the UK’s NHS compared to the Philippines. And as of May 2020, 23% of deaths among frontline workers in the UK were of Filipino heritage, with anecdotes that migrant health workers were being handpicked to work in high COVID-19 wards without adequate PPE.

The impact of COVID-19 is evident among health workers and vulnerable populations in the form of a two-fold impact; on the healthcare workers and the vulnerable populations they serve. For example, in a government-run facility in Illinois, USA, which catered to over 350 adults with developmental disabilities, instances of health workers having only gloves and inappropriate PPE were recorded, putting the lives of these health workers and the vulnerable groups being cared for at risk.

Moreover, Amnesty International has argued that health and essential workers have been exposed, silenced and attacked in the process of raising concerns about these conditions. Intersectional analysis can lead to a radical rethinking of COVID-19 by understanding how certain groups are disadvantaged by multiple intersectional factors that are structural in nature. However, for intersectional analysis to be undertaken, indicators and their associated data should be accessible. New data pedagogies are needed that prioritise intersectionality, strong data governance and community trust.

The need for intersectional COVID-19 data

Research by the International Council of Nursing (ICN) during the COVID-19 pandemic has shown that governments are not systematically collecting and reporting on COVID-19 infection and mortality in health workers. As at April 11 2020, estimates from the WHO were that approximately 22,000 health workers were infected with COVID-19. The ICN, on the other hand reported that as at May 2020, at least 90,000 health-care workers worldwide are believed to have been infected with COVID-19, they add this figure is an under-estimation as data available to the ICN does not cover every country. Additionally, research published in the International Journal of Infectious Diseases revealed that as at October 2020, a total of 287,157 healthcare worker infections and 2,721 of deaths which were COVID-19 related were reported from a survey of 37 nations. This survey was conducted by individual researchers who reached out to members of the Infectious Diseases International Research Initiative for participation, and suggests, again, that the data we can access is subject to different standards and methodologies. This has serious implications for policy making because the lack of standardised data hides health vulnerabilities that may be exacerbated by intersectional factors.

This inconsistency in data standards can be highlighted by a Canadian example. COVID-19 data collection standards vary per province. For example, Ontario uses the integrated Public Health Information System (iPHIS) for reporting and surveillance of diseases. The system is slated to be replaced, since results are sent to health units by fax and infection data per patient is manually entered into the system. Gender, age and health worker related occupation data is collected in Ontario, but this is not the case in Manitoba where partially disaggregated health worker data by occupation is available,  but no age and gender disaggregation is available. Also, in Ontario, health worker related data on occupation included limited categories; health care worker, first responder, doctor, nurse, and dentist. This means that of the 5815 COVID-19 infection cases reported among health workers, 4000 cases were classified as ‘unspecified healthcare worker occupation’. Data for dental hygienist, midwife, other medical technicians, personal support worker, and respiratory therapist were only added to iPHIS on 29 May 2020. The different provincial standards, late reporting and data categories that are not fully disaggregated, lead to chronic underreporting of COVID-19 infection data among health workers and reduces the degree of intersectional data available and our confidence in it.


Underreported data has implications for policy making and exacerbates the continued perpetuation of inequity as it relates to female, minority and other marginalized health workers. Data availability would enable socio-economic research highlighting the linkages between infection and mortality rates through an intersectional lens, allowing researchers to identify and address vulnerability factors. Bringing intersecting disadvantages to the fore can ensure a more equitable and effective response to COVID-19 and future pandemic threats.

During world health worker week, we take a moment to honour those health workers lost to COVID-19, and recommend three resources that can improve the conditions of work for healthcare workers globally:

  • The WHO should continue to build on its expertise in intersectional gender analysis, and lead the development of intersectional databases that enhance the systematic collection of data on health worker infection and mortality.
  • Governments must support and advance the work of the WHO’s Gender Equity Hub and their call for large-scale gender-transformative progress to address gender inequities and biases in the health and social workforce for the SDGs.
  • We should all support and advance the WHO’s campaign to invest in health workers for shared dividends in health, jobs, economic opportunity and equity.


Tobi Ogundele and Margaret Walton-Roberts are from the Balsillie School of International Affairs and members of the Gender and COVID-19 working group Healthcare Worker Sub-Group. Image: “Covid-19 San Salvatore 09.jpg” by Alberto Giuliani is licensed under CC BY-SA 4.0.


Gender Working Group

We meet online every month to discuss key issues, activities, opportunities and ideas for collaboration. We have a long and growing list of resources on gender and public health emergencies.



Gender Working Group

We meet online every month to discuss key issues, activities, opportunities and ideas for collaboration. We have a long and growing list of resources on gender and public health emergencies.