Table 2 Summary table of terms used in Fig. 1 with illustrative participant quotes

1

Trust

Trust is an perception or attitude one holds regarding another actor’s trustworthiness in their behaviour; here ‘trustworthiness’ being the honesty and integrity of the actor’s actual behaviour. This trust may be built on relationships in which both parties have proven their trustworthiness or, in other cases, may be blind trust in those with the power over the data. Regardless, trust suggest that one party is vulnerable and dependent upon others to act in the right manner according to the particular situation, including taking the trustee’s interest into account (McLeod, 2015). In the context of WGS for TB surveillance, issues around trust (or lack thereof) may be present in many relationships including, but not limited to: 1) between health care workers and laboratory staff, 2) the public and the medical community and 3) for concerns of data protection for both privacy and confidentiality purposes, as well as academic.

Well, I think, ultimately, that is what happens, right, is that sort of trust is like – if the folks who are the expert in the technology are saying that these are people who are likely related or that this is like – these are definitely a cluster, then I think you do kind of have to take that on at face value. (Interview #14, Governance). It takes ages to build [trust], and if you make a misstep, the whole thing could be gone. So you have to just be very careful, very open, never to make it seem like you’re on the make. Just be a – and actually, it’s not that hard. You just be a, you know, decent, honest person, and keep people in the loop. (Interview #16, Lab)

2

Data sharing and profitting from surveillance

Fear of the potential that profits may be derived from data acquired through surveillance activities. In conjunction with data sharing concerns previously detailed in the literature, this encompasses the fear that if data is shared with international surveillance databases, the data can then be used to developed new diagnostics and/or drugs that will then be sold back to countries at high, potentially unaffordable, prices. Fear of the potential that profits may be derived from data acquired through surveillance activities. In conjunction with data sharing concerns previously detailed in the literature, this encompasses the fear that if data is shared with international surveillance databases, the data can then be used to developed new diagnostics and/or drugs that will then be sold back to countries at high, potentially unaffordable, prices.

They would – they’re pretty hesitant and what has been done in the past is that if they have volunteered to provide data that companies can come in, identify potential mutations that would be of relevance and make a diagnostic and then sell it back to the country. So the country is a little bit concerned that what may happen is that a company can basically make money off of their data that they’re providing for free. (Interview #1, Lab)

3

Access and Who Benefits

Which regions (e.g. LMICs vs. HICs) are actually able to implement WGS technology - including the sequencing itself and the data analysis - and how we avoid situations where these lower-resourced settings are being ‘used’ by other researchers who derive benefit in the form of papers/scientific stature, while the sequences and the resulting data may not ever feed back into meaningful change in TB policy and practice in the study setting. Which regions (e.g. LMICs vs. HICs) are actually able to implement WGS technology - including the sequencing itself and the data analysis - and how we avoid situations where these lower-resourced settings are being ‘used’ by other researchers who derive benefit in the form of papers/scientific stature, while the sequences and the resulting data may not ever feed back into meaningful change in TB policy and practice in the study setting.

This is all about helping low/middle income countries to get – basically, you know, why should they not be able to reap the benefits of this new technology (Interview #2, Lab).

4

Power of Public Health

Actors within the framework of public health can have immense powers which, while working in the public interest, could result in the undermining of various rights and civil liberties.

Certain things sell, and I think in some ways public health sells, and I just think we need to really make sure that we’re rigorous and when we say something it is of public health benefit - because public health also comes with enormous powers, right - I can look at data, I can look left and let people do things, sometimes you actually, you know basically you have legislation that says you can breach patient confidentiality in the interest of public health - there’s a lot of power associated with public health which means that we have to be absolutely rigorous about that fact that we only use those powers with discretion and we only use it for the purpose for which it was intended. (Interview #10, Governance)

Emerging concerns in relation to whole genome sequencing, flowing from and mapping back to issues of trust

Bioinformatic challenges

5

Utility

In the context of WGS for TB surveillance, utility is the usefulness or benefit WGS contributes to TB programs both nationally and globally for surveillance, diagnostic and TB care purposes. The utility of WGS within individual TB programs appears to depend on a number of factors including (but not limited to): access to the technology, bioinformatic and human analytic capabilities and engagement with public health workers. The utility of WGS appears to be strongest when it is supported with strong epidemiological data.

And so, even with whole genome sequencing, like the discriminatory power, it’s just like not that amazing. So, it really, like what it tells you changes so much from disease to disease, it can be really informative in some situations or really not informative in others. (Interview #8, Public Health)

6

Analytics

Analytics encompasses many bioinformatic challenges that arise from the use of WGS in TB surveillance and care, including: 1) analytic capabilities of both the technology and the professionals working with it; and 2) standardized processes for standard outputs.

Whole genome sequencing offers a whole different - It’s a whole different ballgame in terms of resolution power. That said, at this point in time, it’s not fully standardized on how you do the analysis and that’s part of the problem…. it takes you know a while to develop. So right now for whole genome sequencing in its infancy where yes sequencing is now cheap relatively, you can get a pretty good throughput, but you get a ton of information that comes out of it and it’s how you analyze that information and depending on what programs you use to analyze it, you may get different results and therein lies part of the problem. (Interview #9, Lab)

7

Capacity

Capacity is the ability of countries to actually implement WGS technology into their local TB programs. The high cost of the machine, the sequencing and/or the analytic software may pose challenges to countries, either alone or in conjunction.

But it’s then having a pipeline that can actually interrogate that sequence and develop an interpretation of what that sequence means in terms of the resistance in specific genes that we know are associated with resistance. So, you know, we have the platforms to perform whole genome sequencing, but then a pipeline to actually go from the sample through to sequencing to then interrogating that sequence to an interpretation is the bigger challenge. And this is where countries will need a lot more support. (Interview #4, Lab).

8

Understanding Science

WGS is a rapidly advancing technology that is often progressing at a rate faster than those working with the technology (either indirectly or directly) are able to adapt. As a result, epidemiologists and other public health workers (i.e.. clinicians and nurses) are often left reliant on the lab staff and others with the niched knowledge of the technology to analyze and interpret the output data. This has the potential to have negative impacts on the overall utility of WGS in public health practice.

And so, I think as a result probably some risk that the data from whole genome sequencing phylogenetic analysis that we’re providing the many state and local programs in support of their daily practice, is susceptible to misinterpretation just by virtue of the fact that the workforce capacity in terms of training and education has not caught up yet. (Interview #13, Public Health)

Data protection and stewardship regarding metadata

9

Metadata

Metadata is the individual patient-level data, including administrative data and health records. Question arise regarding whether or not this data should be linked to sequenced genomic isolates being collected for surveillance purposes and who should have access to this data.

So what does it mean if we start, you know, that slippery slope, well what happens if we start sharing like well let’s just share gender now, oh let’s share country of origin now. (Interview #7, Public Health).

a. Linked Metadata

When metadata databases are linked to the sequenced genomic data produced through WGS, it becomes much more powerful tool for tracking outbreak transmission patterns, at risk case identification and potential diagnostic developments. However, it also raises many questions and concerns regarding privacy and confidentiality of patient level data, as well as data protection responsibilities.

But we still have to go through our privacy before we can deposit information and one thing that still seems to be a sticking point, not necessarily just with us, I think with other institutions is any associated metadata. So you know I can deposit TB study number one … A sequence, no problem, they know it’s from [province x], but that’s it. But like I said before what’s going to provide the most useful information is all that metadata which will give us context and I think that we’re not there yet in terms of how do we share this information, because really that’s going to be what’s needed. We need worldwide information, so we have an understanding of all the different environments and the different settings where … You know where we see certain drug resistance, mutations versus what we see in our population or you know new like strains, clusters, that kind of thing and have that understanding. (Interview #9, Lab)

b. Unlinked Metadata

Not linking at least some metadata to the genomic isolates essentially renders the sequenced genomic data produced through WGS useless from a public health perspective as traditional epidemiological activities, such as contact tracing, would still be required to make the data actionable for public health actors.

So now how do we compare them, so we’re going to need to get a lot of information with respect to you know metadata for example that needs to accompany the whole genome sequence data, like, you know, you need the clinical data, the epi data, you know outcome data and that’s going to shape the knowledge that will come out of whole genome sequencing. (Interview #9, Lab)

Consequences and/or implications to data protection responsibilities

10

Stigma

Communities affected by TB tend to already be among marginalized communities. Concerns of further marginalization and stigma arise from the use of WGS for TB surveillance, especially when metadata is linked to the sequenced genomic data. Some of these concerns include: 1) being discriminated against if their identity is discovered through genomic data; 2) public misinterpretation from the media in the reporting of disease outbreaks; and 3) issues of discrimination regarding immigration and care seeking behaviours.

You know, if we identify sort of questions of transmission – TB, for example – you know, are affecting sort of a particular community or population. What may be the implications of that, how would that potentially perpetuate stigma? I mean, that’s sort of been some of my concerns around how the data’s been used for HIV. And so one of my concerns with the process is that I don’t think, like you say, the ethical considerations are really thought of up front. (Interview #14, Governance)

11

Community Engagement

In Canada, the USA and England, TB tends to be disproportionately concentrated within migrant communities, particularly those newly immigrated, and other communities experiencing poverty (i.e.. homeless populations). In Canada, TB is also highly prevalent within Indigenous and First Nations communities. Engagement and transparent communication with these affected communities when reporting out disease outbreaks can help to mitigate some stigma and help ensure that any reports are culturally appropriate and sensitive to the needs of the community.

And then to communicate with the community, I think we need to have community engagement upfront when we start developing communication pathways. So like now, we need to engage advocacy groups, First Nations groups, representatives of different patient groups, to make sure that their thoughts and concerns are well-represented, and then continue to have them on board when we start communicating these results within communities and between healthcare providers and patients. (Interview #11, Public Health)

Challenges assocaited with implementing TB surveillance systems using WGS

12

Implementation/Roll-out

Successful implementation/roll-out of WGS into TB programs for surveillance and care requires that countries have the capacity to implement it, proper analytic systems in place, individuals that understand the science of the technology and the interpretation and finally, buy-in from health care workers that can appropriately communicate it to patients. Additionally, successful implementation of WGS technology depends on: 1) buy-in from state-level and federal administrations; 2) transparency and clear communication strategies for the public and other jurisdictions; and 3) appropriate training systems in place for new professionals working with the technology.

We don’t get much out of it, it doesn’t seem to make sense, the predictive value is poor, we don’t understand what it’s all about, we don’t use them, we ignore them. We think, oh, all this money is going into typing, and then the end-user totally ignores it, it’s pointless. So it’s really important to explain to them the potential value of using the whole-genome sequence data, and allowing them to feel excited about it, and some ownership around it, and hope that that would mean that they engage with it, and then use it, because otherwise it’s a waste of money. (Interview #16, Lab).

13

Jurisdictional Issues

Given greater global connections, TB cases are seldom restricted to one geographic region. Consequently, issues of jurisdiction may occur at the local, state, federal or international level. Issues of jurisdiction may include (but are not limited to): 1) differences in analytic capacity between jurisdiction; 2) ‘ownership’ over the case and responsibility for investigation; 3) communication across jurisdictions and; 4) potential impacts on affected communities including stigma and/or immigration concerns. Consideration of these issues will impact how successfully WGS technologies are implemented into TB surveillance programs.

But how TB programs are organized, I guess we’ll stick to TB, and actually delivered in each province, that’s their, obviously, their purview and it’s not for anyone, including our government, to comment on. (Interview #6, Governance)

14

Communication with non-community stakeholders

Non-community stakeholders includes health care workers (i.e.. clinicians and nurses), laboratory professionals and government officials at state and federal levels. Gaining buy-in and understanding of WGS technology from these stakeholders is vital for successful implementation of this technology. Appropriate messaging of outbreak transmission patterns and WGS technology from TB stakeholders to affected communities should also occur as a result of this.

This new system is quite visual, so if you present the data in terms of attractive-looking trees, or network diagrams, and explain that to them [TB nurses], then they tend to get it relatively quickly, and enjoy looking at it, and it – I think they get enjoyment out of it, because it makes sense. (Interview #16, Lab)