The number of daily Covid19 infections in the UK hit almost 3,000 today. This time last month the number of daily infections were less than 1,000, a month before that they were at less than 500. It looks like we are well and truly heading into the second wave. There is a lot of talk about the possibility of another lockdown and the impact that could have on the economy. It doesn’t have to be that way, other countries, including South Korea and Taiwan managed to get their infections down and keep them down without a lockdown. What they did have however is contact tracing systems that worked. Over the last few months, I have been thinking a lot about contact tracing and how to make contact tracing work.

I have been part of a group evaluating a community contact tracing pilot. The pilot project took place earlier this year, before plans for a national contact tracing system had been announced. I would love to say that now that a national contact tracing is up and running the insights that I captured about how to make contact tracing work are no longer needed. However the recent increase in infections and extensive criticism of the UK contact tracing system demonstrates that this is not the case. The main focus of this blog post is lessons about how to do contact tracing well, based on my evaluation. Some more thoughts of mine are outlined in a piece I co-authored in the BMJ here and the full evaluation report can be found online here

As indicated by one of the volunteers interviewed, one of the main lessons learned was that trust is essential to make contact tracing work.

“you need to build trust and you need to be able to talk to them

(index cases and their contacts),

in a way through which they are willing to offer updated information”

Volunteers found that local connections and frequent communication helped to develop trust. Local connections helped to gain the trust of people contacted in two different ways. Firstly, volunteer contact tracers found that being able to talk about shared experiences of the local area helped to put people contacted at ease. Secondly, being able to provide practical support, such as being able to refer people to support offered locally helped.

Contacts were more likely to offer cooperation if they felt that the contact tracers were really trying to help them and were on their side. Regular, daily communication was found to be effective as trust was developed over time. This may seem counterintuitive given the urgency of getting in contact with people to get them to self isolate, but the community contact tracers often found that the people they spoke to only agreed to pass on their contacts after regular communication had been developed over several days.

In a previous blog post I highlighted some of the reasons why the approach being taken by the UK government was not instilling trust in people. I highlighted issues with the private companies commissioned to conduct contact tracing. Issues that detracted from trust included a data breach by Serco at the beginning of the contact tracing process and alleged links between the company commissioned between the company commissioned to make the app and Cambridge Analytica. For more detail, see the earlier blog post here.

In the contact tracing pilot I evaluated links to local community groups were also found to be helpful in developing trust and getting people to self isolate. This linked to another finding, the impact of the project was not just about getting people to self-isolate but also the practical and emotional support that was provided. People who were self isolating often felt isolated, with little contact from other people. People contacted stated that it felt good to have someone checking to make sure that they were Ok every day. Contact tracers stated:

“He felt that he wasn’t alone with me with me, communicating. “

and

“after two weeks she was saying how much she appreciated me calling her son.”

Practical support was just as important. Sometimes for people to be able to self isolate they needed to tap into support networks, for practical help such as having someone to do food shopping or to pick up medications. It appeared that a holistic approach focussed both on providing support and getting people to self isolated seemed to be effective.

Volunteers were skeptical of the potential of people calling from a call center to develop trust, one stated:

“if you’ve got someone from a call centre ringing you up and saying

you must lock down for 14 days, I just don’t think it’s going to work”

Another failing that has been highlighted about the national system is lack of relevant skills and expertise of many of the contact tracing employed by Serco.  At first glance it may appear contrary to suggest that a volunteer based approach can harness a higher skill set than one that employs people to conduct contact tracing.  In this instance, that is what was found.  It was strongly emphasised that contact tracing cannot be done by just anyone, and most of the volunteer contact tracers in the pilot project brought with them a significant existing skill set.  One stated that:

“If I am being frank I was in the main drawing on a skill set that I already had”

Most had previously worked in health or care roles, some at quite a senior level.  In addition the management committee of the project was primarily composed of retired healthcare professionals who were able to mentor volunteers to develop the expertise required.  The importance of the expertise of the steering group was highlighted by the following volunteer quote:

“I wouldn’t have got involved with any Tom, Dick or Harry

who decided to set up a community contact tracing and group”

The support structures that were put in place were found to contribute to the success of the pilot.  These structures provided emotional as well as practical support, as the process of contact tracing can be emotionally draining as well as time consuming.  Structures put in place also enabled a continual process or review and learning, to help address new and unexpected challenges.  Our related recommendations are that contact tracers will have relevant skills, support structures are put in place to provide emotional support, to provide access to additional expertise and to aid in the development of volunteers and that learning is captured to continually inform and prove processes.  

Cooperation from employers was found to be a significant barrier.  I suggest that community organisations need to work in collaboration with official structures to provide the legitimacy needed to gain access to contacts in workplace settings.  As more people stop working remotely and get back to working in close contact with their colleagues such cooperation will become increasingly important.  

An issue related to employer cooperation is employee cooperation. This was another issue that came up in the evaluation. Some people who are asked to self isolate may feel that they cannot afford to do so. This is particularly an issue for people who are self-employed, low paid or on zero hours contracts. Whilst the £13 a day now offered to people self-isolating may be a step in the right direction, as outlined in this article , £13 a day has been criticised as a slap in the face, and not enough to convince some people to self isolate, some may simply find that it is not enough to pay the bills.

There is also the issue of apps. Two volunteer, contact tracers discussed the potential integration of apps into a community contact tracing project. Apps have the potential to provide a lot of data about the people that those infected with Covid19 have been in contact with. Apps have the potential not just to provide this data faster, but also to simply provide more data. Even if contacts are trying to be cooperative they might not know or remember everyone they have been in contact with. As more people start going back to work, to school or universities keeping track of all contacts could become more challenging.

Simply deferring to technology, however, I think, is also problematic, it is problematic because of the issues of trust outlined earlier. This article found that app based contact tracing systems around the world have been significantly hampered by the issue of lack of trust. One way to get around the issue of trust I suggest is to integrate apps into a community focussed approach. Within this I suggest the contact should be made by a human, ideally someone local who is able to provide or signpost people contacted onto practical and emotional support.

Finally reaching people with symptoms and their contacts as soon as possible is very important.  Networks can be developed to reach people sooner.  The pilot found that waiting for post test referrals led to too much delay.  I suggest that there may be value in starting contact tracing as soon as people have symptoms rather than waiting for positive test results to appear.  

The full evaluation report is available online here.