Background

The COVID-19 pandemic caught the world unprepared to respond in a structured way, despite warnings following previous public health emergencies of international concern (PHEICs) [1] [Severe acute respiratory syndrome (SARS-CoV 2002–2004), Middle East Respiratory Syndrome (MERS-CoV 2012–2015), Zika (2015–2016), H1N1 influenza pandemic (2009), Ebola (West African outbreak 2013–2015, outbreak in Democratic Republic of Congo 2018–2020)].

The COVID-19 pandemic triggered one of the largest global health crises in more than a century [2]. Even the world's most developed economies faced the consequences of the pandemic utterly unprepared. As of 17 May 2023, around 766.4 million people were infected with this deadly virus (SARS-CoV-2) and 6.9 million deaths had been attributed to it [WHO Coronavirus (COVID-19) Dashboard, https://covid19.who.int].

Some countries managed to contain the virus, but for most, the responses to the pandemic ranged from poor to disastrous [3].

The COVID-19 pandemic represented an additional stressor to health and care systems, aggravating already fragile systems and undermining progress towards the sustainable development goals (SDG) [4]. The pandemic specifically impacted negatively on progress towards achievement of health-related SDGs [5], in particular, target 8 of SDG 3, achievement of Universal Health Coverage (UHC), including financial risk protection, access to quality essential healthcare services and access to safe, effective, quality and affordable essential medicines and vaccines for all (https://www.un.org/sustainabledevelopment/health/). This disruption to healthcare services is greatest among low-income countries [6].

Health and care workers (HCWs) are part of the first line response facing the pandemic. Nevertheless, as there are no official global figures, casualties are likely to include a considerable sum of HCWs [3]. The pandemic indisputably has affected their personal/family life, their working conditions, their physical, mental and social well-being, and they have been documented to have a higher risk of morbidity and mortality associated with SARS-CoV-2 than the general population [7]. The feminization of the workforce is a reality (globally, around two-thirds of the total HCWs are women), and the pandemic has brought women an increasing burden of care, unpaid care and increased domestic responsibilities [8], worsened in a context of wage gaps between men and women in the health sector, as well as reported violence against women as a symptom of power imbalances [9].

As a result, HCWs in several countries resorted to industrial actions, protests, strikes and lockouts (IAPSLs). When trying to address these IAPSLs, policymakers and managers faced a vacuum of knowledge as the evidence produced from other recent PHEICs has hardly been reviewed systematically. Previous efforts to synthesize evidence and provide policy guidance to inform the decision process have been carried out by the World Health Organization (WHO) to guide “human resources for health managers and policymakers at national, subnational and facility levels to design, manage and preserve the workforce necessary to manage the COVID-19 pandemic and maintain essential health services” [10]. However, as the pandemic advances and new developments are in place (i.e., mass vaccination, new strains of the virus) knowledge needs to be updated and made available.

HCWs’ IAPSLs are an ongoing phenomenon both in the Global North as well as in the Global South. The most reported in the literature are strikes, with all the ethical issues associated with measures taken by HCWs that hamper HCWs’ “duty to patients” and deprive the population of the care they need or seek [11,12,13].

Apparently, there was a substantial uptick in strike actions by HCWs in several countries at some point during the pandemic (e.g., Hong Kong, Zimbabwe, United States, South Korea, Kenya, Spain, Bosnia, Peru, Myanmar) [3].

The general expectation is that HCWs’ strikes would lead to a decline in care and increase in mortality. However, several studies have suggested that when doctors go on strike, mortality rates paradoxically may fall [14].

This systematic review [15] was designed and conducted to answer to the review questions in Box 1. In this paper, we report the baseline results of the systematic review.

So that the article would not lose its relevance, the authors made a non-systematized update to the search, repeating the search expression, covering the period between March 1, 2022 and January 31, 2024. Relevant articles were added to the discussion of results.

Methods

This systematic review followed the PICOC (Population, Intervention/phenomenon of interest, Comparison, Outcomes, Context) structure proposed by Petticrew and Roberts [16] (Additional file 1).

The protocol for this systematic review has been registered in PROSPERO (registration PROSPERO 2022 CRD42022324115 Available from: https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42022324115.

Eligibility criteria

The review included studies in which participants were HCWs participating in IAPSLs in the context of the ongoing COVID-19 pandemic or other five PHEICs since 2000 (SARS-CoV from 2002 to 2004, MERS-CoV from 2012 to 2015, Zika from 2015 to 2016, H1N1 influenza pandemic during 2009, West African Ebola outbreak from 2013 to 2015, and the Ebola outbreak in the Democratic Republic of Congo from 2018 to 2020), the intervention/phenomenon of interest. Whenever mentioned comparisons related to no-intervention or other interventions were considered. Outcomes considered relevant included the impact on service delivery (e.g., service disruptions, lockouts, days of work lost, numbers and type of HCW involved, lost working days), in-patient morbidity and mortality and the interventions to address HCWs’ demands.

HCWs are defined as all health-related occupations in the health and social sectors of employment, and various settings of employment, including health facilities offering all level of care, institutional and residential care homes, amongst others [17, 18].

IAPSLs are defined as any collective withdrawal of services or work stoppage, temporary show of dissatisfaction by employees to protest against bad working conditions or low pay and to increase bargaining power with the employer and intended to force the employer to improve them by reducing productivity in a workplace, by a group of individuals that work in health facilities offering all level of care (including those involved in patient care as physicians, nurses, midwives, and pharmacists), institutional and residential care homes, amongst others [19].

A PHEIC is defined in the International Health Regulations (2005) as “an extraordinary event which is determined to constitute a public health risk to other States through the international spread of disease and to potentially requires a coordinated international response” [20].

The types of publications elected were observational (i.e., cohort and cross-sectional), experimental, quasi-experimental, reviews, mixed methods and qualitative studies published between 1st of January 2000 and 1st of March 2022 in English, French, Hindi, Portuguese, Italian or Spanish.

Information sources and search strategy

For this review, the following scientific databases were used: PubMed, Embase, Scopus, LILACS via BVS, WHO’s COVID-19 Research Database, International Labour Organization (ILO) [21], Economic Co-operation and Development (OECD) [22] and The Health System Response Monitor (HSRM) [23] and Google Scholar.

To identify the search terms, the controlled vocabularies of the health area DeCs (Descriptors in Health Sciences), MeSH (Medical Subject Headings) and Emtree (Embase Subject Headings) were consulted. The search strategy is detailed in additional file 2.

Selection of studies

We used Endnote [24] to collect, organize and manage references retrieved from the searches of the different database. Once this phase was completed, we uploaded the references file to Rayyan [25].

The selection using the eligibility criteria was made by two researchers independently. Differences were discussed with a third researcher until a consensus was reached.

Eligibility and exclusion criteria were first applied to titles and abstracts of the documents uploaded in Rayyan, independently by two reviewers. In case no abstract was available, the criteria were applied either to the executive summary or to the first five pages of the introduction of the document.

If eligibility criteria were met, documents were selected for full-text analysis, applying again the eligibility criteria (Fig. 1). If again fulfilled, relevant data were then extracted from the document. In case of disagreement a discussion between the two reviewers was done and a third referee reviewer was consulted. Inter-reviewer agreement was measured by estimating Kappa statistics and computing sensibility and sensitivity was computed to assess and the quality of this process [26].

Fig. 1
figure 1

Eligibility criteria flow diagram

Full size image

Inclusion of publications is described as recommended by PRISMA [27, 28] (Fig. 2).

Fig. 2
figure 2

PRISMA flow chart of studies selection phase

Full size image

Data extraction

Data extraction from included studies was randomly distributed between the two reviewers and was performed using an electronic form in REDCap [29]. Information extracted from selected articles included author, year, country, context, title, language, study design, and sample (Table 3—Data extraction).

Assessment of the risk of bias of the studies

To assess the certainty of the evidence of included studies, we used an electronic form of the GRADE CERQual—Confidence in Evidence from Reviews of Qualitative research) for quantitative and qualitative studies [30] developed in REDCap and specific for the study design (Additional file 3).

The GRADE CERQual includes four components to assess the degree of confidence in the results of reviews (also called qualitative evidence synthesis): methodological limitations, coherence, data adequacy and relevance. This tool should not be seen as a mechanistic approach as confidence assessment is a subjective process [30].

Data synthesis

Narrative synthesis was used to review and synthesize the extracted data from the systematic review. We undertook a qualitative content analysis, organizing the dimensions interconnected with the research questions, as follows: the nature and intensity of the IAPSLs (related with review question 2); the main reasons leading to HCWs’ IAPSLs (related review question 3); the results of these IAPSLs—outputs, outcomes, and impacts (related review questions 1, 2 and 4). In the analysis, there was a concern to verify the gender issues addressed in the articles.

Table 1 Number of publications from the first phase of the systematic review
Full size table
Table 2 Inter-reviewer agreement between the two researchers
Full size table
Table 3 Data extraction about included publications
Full size table
Table 4 GRADE CERQual assessment
Full size table
Table 5 Geographical dimension and duration of the strikes
Full size table
Table 6 Drivers of HCWs strikes during public health emergencies of international concern (PHEICs)
Full size table

In total, 493 studies from databases were identified. After the exclusion of duplicates (n = 26), 467 references remained, and 1163 publications were identified from grey literature but only the first 100 publications from each information source were screened (n = 400). First, after reading the title and abstract, 102 studies were select for retrieval (98 from databases and 4 from grey literature). Then, 91 studies were selected for full text reading (87 from databases and 4 from grey literature). We identified 18 publications, all in English, which met the inclusion and quality criteria for data extraction (14 studies and 4 other publications). The entire selection process is documented in Fig. 2.

The summary of the total number of articles, conflicts between the 2 researchers for sensibility test, reasons for exclusions and included articles is presented in Table 1.

The inter-reviewer agreement, according to Kappa standard, was calculated and is shown in Table 2.

Results

1656 records were retrieved, duplicates were checked, and the screening inclusion criteria were successively applied (in the different phases as shown in the flowchart), until the final phase in which we were left with 91 documents that were selected for full text screening (87 from the databases data and 4 studies identified via other methods). We included 18 publications after assessment for eligibility, ranging from 2007 to 2022. Ten were cross sectional studies, 4 were qualitative studies and 4 were reviews.

Fourteen country-specific publications met the inclusion and quality criteria for data: 11 were from Africa (7 from Kenya, and 4 from Nigeria); 2 from Asia (South Korea), 1 from Europe (Croatia). Four review articles were also included: 1 focused on several low-income countries, 1 from South Africa, 1 from India, and one with a focus on the United States of America (USA). Most of the publications referred to IAPSLs by physicians and/or nurses (n = 12).

Details on data extracted from the publications included in the review are presented in Table 3.

The findings from the records, outcomes and details of GRADE’s assessment are shown in Table 4. The evidence in the studies ranged from low to high confidence. The majority (14/18) of publications were rated high and/or moderate, based on details provided in the summary assessment, the findings should be with confidence.

Nature and intensity of the industrial actions, protests, strikes and lockouts

The majority of IAPSLs addressed in the literature reviewed referred to HCW strikes. Only in two studies, the HCW protests are entitled industrial actions. Given the nature of the IAPSLs addressed in the literature review from now on we will refer to strikes rather than IAPSLs when describing our findings.

Strikes occurred in two contexts: countries where there is an entrenched environment of IAPSLs by HCWs, which has been aggravated by COVID-19 and other PHEICs; and countries where HCWs’ strikes are unusual.

Countries with an entrenched environment of strikes by HCWs, aggravated by COVID-19 and other PHEICs, include India [32] and South Korea [45, 46] in Asia and Kenya [33,34,35,36,37,38,39], Nigeria [40,41,42,43] and South Africa [44] in sub-Saharan Africa. In Nigeria, in the last decade, no sector has been more affected by strikes than the health sector and HCW industrial action is described as “commonplace throughout history” [42]. In South Africa, HCWs’ strikes has been particularly associated to strong nurses' activism [44].

There is a growing labour unrest also in countries not widely perceived to experience this problem [49, 50].

The most common strike modality was the one engaging more than one type of HCWs (9/18 publications) [34,35,36,37,38,39, 41, 42, 48], followed by strikes by physicians (6/18) [31, 33, 40, 43, 45, 46] by nurses in South Africa and in the USA (2/18) [44, 47] and by community health workers in India (1/18) [32] (Table 5).

Both the duration of the strikes and the geographical dimension are not homogeneous. Strikes’ duration ranged from 2 days to 36 months. In some countries, such as Nigeria and Kenya, they tend to have a recurrent nature. They could be restricted to one institution, one locality, or extend to several locations, state-wide, to several states or even countrywide (Table 5).

The literature refers mainly to Asian and African, and it is striking the absence of quality literature on the issue from other parts of the World.

Main reasons leading to HCWs’ industrial action, protests, strikes and lockouts

In this systematic review, we intended to describe the main grievances of HCWs related to one or more of WHO’s health system building blocks [51]. Their grievances are mostly concerned with five building blocks: leadership and governance, financing, HCWs, medical products and technologies and service provision.

Many HCWs work under health and care systems with significant leadership and governance fragilities, inadequate payment systems, under-resourced with health personnel, with inadequate access to medical products and technologies (ventilators, oxygen, and personal protection equipment (PPE), providing unsafe care in sub-optimal conditions. These conditions are further aggravated during PHEICs, unprecedented by all standards but completely predictable [1]. These can be considered the main drivers of HCWs’ strikes during PHEICs (Table 6). As we can see in Fig. 3, issues related to leadership and governance and medical products and technologies were the ones that came up most frequently in the articles analyzed.

Fig. 3
figure 3

Leading causes of strikes

Full size image

When we compare the pre-COVID-19 strikes to those that occurred during the pandemic, the protests have in common demands for better wages and payments as well as complaints about leadership decisions and contestation over healthcare policy options. The reference to the lack of PPE and harassment by security forces due to the lockdown is only mentioned in strikes during COVID-19 and strikes in this pandemic period had a greater focus on unsafe working conditions. Despite the specificities of demands of HCWs during PHEIC the common thread in all protests is related to the eventual failure of governments to maintain an operational health system and to provide HCWs with adequate means to carry out their work [52].

During the COVID-19 pandemic, the extraordinary challenges governments and health and care systems faced in creating safe environments for HCWs has also led to a spate of protests from nursing organizations throughout the world. Mavis Mulaudzi et al. [41] mentioned that the American Nurses Association, the Royal College of Nursing, and the International Council of Nurses have all registered their dissatisfaction with the limited supply of PPE and the subsequent risk this poses to nurses as health workers involved in clinical care delivery. They also refer that African nurse unions united in collective protests against inadequate working conditions, defending that African HCWs must raise their voices regarding their needs, including the provision of PPE, to inform policies during the COVID-19 pandemic [41]. “An independent analysis has identified industrial dispute and strike action in 84 countries since February 2020; of which 38% and 29% of strikes are due to poor working conditions and lack of PPE, respectively” [53].

PHEICs created a ground propitious to HCWs express their demands and manifest their protest over long-standing fragilities in several of the building blocks of the health and care systems. One example is Nigeria where one of the causes of the strikes during Ebola related to leadership and management: perceived deficiencies in the health system may have been exacerbated by the PHEIC. This is also the case for Kenya and South Korea, where the reasons for some of the IAPSLs do not seem directly linked to the concurrent PHEIC that may have served simply as a backdrop to continue protests related to long-standing grievances [39, 46].

In Kenya, there was also a different outcome for physicians’ and nurses’ demands following the 2013 strikes, which led to subsequent disputes related to promises made to nurses. This indicates that, in the context of labour disputes, if there is agreement between the parties that are perceived not to have been fulfilled, this creates an environment favorable to subsequent episodes of strike and protests, which when coincident with PHEICs, exponentiate the risks for all, patients and HCWs.

Both Asian and African countries seem to live in an environment conducive to the appearance of strikes during PHEICs, which only aggravate work conditions that were already difficult, where HCWs already faced a variety of challenges to pursue their professions.

HCWs’ industrial actions, protests, strikes and lockouts—outputs, outcomes, and impacts.

Results derived from strikes may be analysed from the perspective of the health and care system beneficiaries (the users and patients) (Tables 7 and 8) and from the perspective of the grievances of the HCWs that led to the IAPSLs (Tables 7 and 9).

Table 7 Analysis of outputs, outcomes and impact from two different perspectives
Full size table
Table 8 Results observed during the strike from the perspective of service users
Full size table
Table 9 Results from the perspective of HCWs
Full size table

Table 8 reflects a clear impact on outputs and outcomes of services (particularly on the poorest). It also reflects how difficult it is to measure impact on mortality and morbidity unless researchers take either a population-based approach (as done by Erceg et al. [31] in Croatia or a systems approach as done by Adam et al. [33] in Kenya). Cunningham et al. [14] drew attention to the limitations of the use of mortality as outcome, because death can be considered in some way a rare event, and therefore is not a good indicator for changes that do not culminate in death, but that can increase suffering and delays in the use of health services. Smith et al. [54] argue that despite the argument of risks of strikes to patients, there is no clear evidence on an increase in patient morbidity or mortality during periods of strike action. Meanwhile, Friedman et al. [55] showed, based on a population study, that children born during HCWs strikes that took place in Kenya between 1999 and 2014 are more likely to suffer a neonatal death.

Essex, Brophy and Sriram [56] find that most of the strike action occurs in response to structural failings of health systems such as austerity, underinvestment and de-prioritization of health, occupational hierarchies and intersectional power dynamics in health services, and larger societal issues, that are components of health system resilience and sustainability. In light of the recent industrial action in the UK, authors argue that there are risks in failing to strike, insofar as strike action has the potential to change the trajectory of failing healthcare systems. They challenge the framing of strikes as unquestionably harmful to patients, asking the inverse and often overlooked question—that is, how might failure to strike adversely impact patients? Authors argue that when health workers lack other avenues to voice concerns, the failure to strike may actually be more harmful to health in the long run. Also, there is knock-on effects of strike action, particularly in low-resource settings. For instance, increases mortality in nearby facilities dealing with surges in patient pressures which is a reflection of systemic concerns around health service access.

The results must also be interpreted against a background that the PHEIC context itself may have a significant impact on outputs, outcomes, and impacts, and that the strike is a reaction to these rather than just adding to them.

Table 9 reflects the scarcity of evidence found from the perspective of the results achieved (or not) by HCW due to the strikes undertaken. Only in two publications was possible to gather information on this topic.

Discussion

HCWs’ strikes “create an ethical tension between an obligation to care for current patients (e.g., to provide care and avoid abandonment) and an obligation to better care for future patients by seeking system improvements (e.g., improvements in safety, to access, and in the composition and strength of the health care workforce). This tension is further intensified when the potential benefit of a strike involves occupational self-interest, and the potential risk involves patient harm or death” [57]. But when strikes happen because HCWs protest because salaries are not compatible with the cost of living or the degradation of public health services and underfunding in the health sector, authors such as Smith et al. [54] consider that that there is an ethical concern in the protests. Adobor [58] adds, from a utility-maximizing perspective, that some strikes may be justified if their results mean long-term improvements in the health facilities that serve the majority, being particularly applied in the case of developing countries, considering the fragility of health systems and medical care in these regions, where doctors and other HCWs face the moral dilemma of protecting vulnerable patients or using strikes as a tool to guarantee better health service conditions.

The ethical dimensions of HCWs strikes during COVID-19 was also emphasized in the literature reviewed. An example was the HCWs strike in Hong Kong, where about 8000 HCWs participated in a 5-day strike in early February 2020. Despite the considerable support from public opinion, the strikers were accused of violating professional ethics and of neglecting their accountabilities, which led to moral distress. Evidence indicates that strikers showed care and concern for the safety of the community, sustainability of the health care system, and well-being of all people in Hong Kong [50].

When discussing the ethical dimensions of strikes, the right of populations to receive health care is identified, but it cannot be forgotten that HCWs should be able to work safely and without fear of threats. Thus, without adequate institutional protection, self-protection is justifiable and moral [50].

In this context, Mavis Mulaudzi, et al. [44] critically discussed the ethics of nurses’ choice to strike during the COVID-19 pandemic, and one of the arguments presented is that these HCWs involved in clinical care delivery are most exposed to risks and must be able to express themselves when the conditions for safely performing their work are not guaranteed.

An identical evaluation was made by Murphy [47] about the example of strikes performed by nurses in the USA during COVID-19. The American Nurses Association has published guidelines for working in emergent disasters, including pandemics, based on earlier guidelines by the Institute of Medicine. But a pandemic represents an exceptionality for everyone, including HCWs. The pandemic puts not only patients at risk of infection and possible death, but also HCWs and their families.

According to Murphy [47], what determines the ethics of a strike or work stoppage is the confluence of three factors—the significant risk for HCWs, the duration of this risk, and the ability to mitigate the risk. As a rule, morally, strikes, work stoppages and job reductions by HCWs during disruptive disasters are a wrong choice, as it can put patients at risk. But when rare circumstances are experienced, however, work stoppages can be ethically justified.

The circumstances that may cause a strike in times of pandemic are: HCWs are put at significant risk of infection, damage and even death; the duration of the risk is persistent; inadequate PPE and staff are available to mitigate the risk. Because this means that current patients, future patients, and healthcare and HCWs are placed in an unsustainable and unsafe health care environment [47].

Irimu et al. [59] produced a commentary on HCWs’ strikes that occurred in public health services in Kenya since devolution of healthcare services in 2013. The disproportional effect of the strikes in the poor who are unable to afford private sector alternatives was stressed, raising the ethical dimension of HCWs’ strikes. People's right to health care appears to have been violated to the extent that no public hospital had nearby health services available. This also seems to indicate the need for the specific roles and responsibilities of national and municipal governments to be clearly defined in decentralized health systems.

Kenya is an example of a country where the right to industrial action and strikes was granted but where the institutional mechanisms implemented to manage the labour conflictual environment could be further strengthened. In addition, the specificities of HCWs’ protests in low-income countries with a disproportional impact of the grievances in the poorest population dependent of public health services (ethical dimension of the strike in these cases) should be tackled.

The literature review shows a clear impact on outputs and outcomes of services particularly on the poorest during PHEICs, for instance, when the strikes represent a limitation in terms of access to public health facilities forcing the patients to pay for private services. Similar results were already mentioned as HCWs’ strikes appear to have a greater impact on the poor and vulnerable with no alternative means of obtaining healthcare particularly in countries where provision of healthcare is mostly dependent on public healthcare services [19].

Our review reports that the main reasons for HCWs to undertake a strike are related to the deficiencies of the health systems, which were aggravated in a pandemic context: the significant leadership and governance fragilities of many health and care systems, inappropriate remuneratory systems, health personnel scarcity, inadequate access to medical products and technologies (ventilators, oxygen, and PPE), providing health care in suboptimal working conditions. In India, additionally to the factors mentioned above, the violence against doctors and healthcare professionals that COVID-19 has fueled is mentioned as a motivation for the protests [60].

A study about nurses’ strike in Argentina during COVID-19 [61] also stated that strikes occurred due to work overload and lack of supplies for protection against COVID-19, but the demands exposed historical problems of the profession and more structural problems of the health system (for example, poor working conditions, labor-intensive processes and times, low wages and the consequent permanent lack of personnel to cover the sector's growing demand).

Systematically addressing these challenges, by making adequate investment in education and employment, including decent working conditions, occupational health and safety, fair remuneration, may therefore have the potential to reduce the occurrence of strikes.

These findings have similarities with others previously published pointing as main upstream drivers the rise of consumerism in healthcare, loss of professional autonomy by physicians, many of whom, like other HCWs, work as employees of public or private healthcare organizations with different perspectives of “professionalism” and pressure for unionization [13, 19, 57]. In these organizations pressure for improved performance is escalating “as healthcare institutions attempt to improve the quality of service through restructuring and change, which leads to greater job dissatisfaction, higher turnover, lower morale and increased industrial actions” [62]. Downstream drivers of HCWs’ strikes include failed employer–employee negotiations regarding fair remuneration and benefits, infrastructural deficiencies and inadequate working conditions, concerns about exposure to violence in the workplace and the most diverse policy issues (usually related to workforce development or that hamper the ability of HCWs to carry out their duties effectively) [12, 19, 48]. Interprofessional rivalry for dominance within the hierarchy of healthcare occupations leading the health care system has been cited as a cause of HCWs strikes in Nigeria [19].

Problems in leadership were the cause of protests by HCWs in all the strikes analyzed, those that occurred pre-COVID-19 and during the pandemic. Essentially, there are problems with fulfilling established agreements with health workers, but also a claim to be involved in political decisions and healthcare organization and strategic options. Thus, there seems to be a need to strengthen leadership and management capacities at different levels of health systems (political decision-makers, managers, clinical directors, directors of community health), including through the skills of individual health system leaders and managers, but also in terms of creating or strengthening mechanisms for multi-constituency and intersectoral policy dialogue; processes and mechanisms for addressing labour relations and dispute resolutions may require dedicated attention.

According to several studies [12, 14, 19, 63, 64], the main results of HCWs’ strikes are disruption of healthcare service delivery, leading to cancelation of outpatients' appointments, hospital admissions, and elective procedures and surgeries. Existing evidence suggests that strikes have little impact on in-patient morbidity. There is also no clear evidence of increased in-patients' mortality during strikes, except in isolated cases, where emergency services were also withdrawn during strikes. In some studies readmissions were higher for patients admitted during a strike [33].

The scarcity of population-based analysis to measure impact of HCWs strikes on mortality and morbidity limit the values of these findings as serious morbidity may happen to be transferred to the community rather than taking place in health facilities during strikes.

The literature on the impact of strikes on management reveals “feelings of angst and long hours worked by management”, persisting conflicts and divisions can obstruct teamwork and affect the working environment negatively post-strike. Younger managers, in particular, may need coping support to deal with negative effects on them personally, on their family life and their professional life [10]. In this same direction, Mayaki et al. [41] mentioned the barriers to teamwork including professional hierarchy, role ambiguity, and poor communication as reason to HCWs strikes during PHEICs.

The limited literature suggests the need to invest in measures to prevent strikes, namely respect for direct care providers, a shared governance process, professional autonomy and contribution, and continuous quality improvement [12]. The literature review identified the lack of governance and leadership as one of the reasons to HCWs strikes in Africa and Asia.

In low-income countries, successful resolution of IAPSLs seems easier when “other ministries (finance or public administration ministry) or higher levels of decision-making (such as Prime Minister or President) were involved, rather than the health ministry alone”. Involvement of external international actors in the negotiations is rare, “with the notable exception of human rights nongovernmental organizations (NGOs) in the United Republic of Tanzania in 2012 and Chad in 2018, and the World Bank’s intervention in Guinea Bissau’s health and education workers’ strike” [48].

Despite the recognized feminization of the workforce globally [8], in the review we only found one study [32] that explicitly addressed gender issues as a reason for the strike, the way the pandemic aggravated the vulnerabilities in which women work (low wages, work exposed to violence, poor working conditions and lack of social protection, namely maternity leave). A gender lens is necessary to understand the context, the motivations of health workers' strikes and, above all, the possible need for targeted interventions to meet specific needs.

There are a few limitations in this study. We only identified a limited number of studies from each region, consequently the results might not be generalizable to similar populations in corresponding regions. The reduced amount of literature included in the review did not allow us to fully answer all the initial questions, especially evidence related to the impact of strikes and interventions implemented to respond to the demands of HCWs in strikes.

Final considerations

IAPSLs by HCWs are not new, but the context of the COVID-19 pandemic has reinforced the need for more in-depth understanding of the phenomenon, with more studies in different countries with distinct health and care systems and interprofessional environments.

The rapid upsurge of the COVID-19 pandemic, like previous PHEICs, aggravated unfavorable working conditions of HCWs both in the Global South and the Global North. These seem to be associated with an above average resort to IAPSLs by HCWs in several countries of the world, with a larger concentration in Asia and Africa.

There is a dispersion in the drivers for the HCWs’ IAPSLs during PHEICs. However, grievances related to medical products and technologies (especially lack of PPE), and financing (disputes due to low wages) seem to stand out. Leadership and governance issues also seem to take on particular relevance as motivations for the protests.

The most common form of IAPSLs reported in the literature during PHEICs are public healthcare sector strikes. The review did not detect any publication on IAPSLs affecting the social care sector. Young HCWs are more prone to resort to strikes when compared to more senior HCWs. This is exemplified in Mexico, where medical students organized a protest during COVID-19 instigated by the death of two young physicians in the course of their work, denouncing "the exploitation of medical health care workers in Mexico" [65]. In India where young physicians went on a strike protesting against an attack on resident doctors in the beginning of COVID-19 pandemic [66]. In addition, in Malaysia, newly qualified doctors organized a nationwide strike in 2021 due to the unequal treatment faced by contractual workers, frustration with the government's lack of long-term solutions, and the enormous burden of the COVID-19 pandemic on this group of precarious workers who were on the front line, and more than half suffered from burnout [67].

It is critical to look at how the decision to go for a strike is conceived in response to HCWs grievances. Is it most common in places where formal associations/unions of HCWs are already well established? Or does the intensity and pattern of strikes among HCWs depend on the governance of health and care institutions and on how they deal with grievances and try to address them before they reach a “striking” point? How important is inter-professional dialogue? And many other issues that need attention as they are not clearly addressed in the literature reviewed. These issues must leave no one behind, ensuring that all categories of HCWs, including less visible categories like CHWs, are considered.

But above all, it is necessary to focus on the preparedness of health and care systems to respond adequately when PHEICs appear. Some of the literature reviewed suggests that if the institutions and health and care systems are not ready, future PHEICs will find HCWs already working dissatisfied and/or at their limit, ready to resort to IAPSLs due to the additional pressures to which they will be necessarily exposed to.

Once a strike begins, it is important to look at the success stories of favorable resolution of strikes among HCWs. What actors were involved and what role did they play? What negotiating processes were considered? What grievances were resolved, and which ones persisted? What contextual factors interfered with the negotiating process? None of the literature selected in this LSR have addressed this.

The main impact of strikes is on the disruption of health care services’ provision. The full impact of the strike on health and care services will not be properly understood unless a system-wide approach is taken, rather than an approach limited to the striking institutions. This is important to allow policymakers to address the health and care needs of the population during the strike in a way that minimizes an additional economic burden to the poorest families that can´t afford to pay private health care when IAPSLs hamper the possibility of access to publicly provided health care.

The full impact of strikes on mortality, disability and morbidity of patients and populations is not possible unless the rates at estimated for whole populations rather than just the population served by the striking institutions, as done by most of the literature reviewed.

When HCWs’ strikes occur in the context of fragile health systems they may jeopardize the fundamental right to health and prevents access to health and care services of the poorest and most vulnerable. Some of the publications discuss the ethical dimensions of HCWs’ strikes in several parts of the world. Ethical tensions escalate when the potential benefit of IAPSLs involve professional self-interest of HCWs (e.g., salary and allowances’ hikes, better working conditions) and the potential risk involves patient neglect, harm or death. It is reported that attitudes of HCWs towards IAPSLs and their ethical concerns differ based on their level of training and their career stage. In the context of a PHEIC HCWs may feel trapped between their ethical obligations towards the persons under their care and their right to claim protection for themselves, from the poor working conditions and lack of PPE, and all the life-threatening risks they face.

Finally, the literature reviewed in the initial phase of this systematic review meets the objective of “understanding the impact of HCWs’ IAPSLs related to the COVID-19 pandemic and to other PHEICs since 2000” but adds little to understand “the relevant interventions to address these IAPSLs” reinforcing the important of regular updates, refining the search terms to allow the identification of literature, if available, that helps to close the knowledge gaps identified. No studies were found in the literature on the analysis of political responses to resolve strikes, the position of the different actors involved in strikes, or negotiation processes between the parties to reach eventual agreements.