University College London
Team Lead, Product Designer
3 months
Jackee Hernandez Torres, Ollie Day, Sabrina Zhong and Tianyun Hu.

LongCo is a digital health intervention for people with long-lasting symptoms of Covid-19.

The Challenge

Help patients manage and reduce symptoms

For some people, Covid-19 can cause symptoms that last weeks or months after the infection has passed. This is referred to as post-covid syndrome or "Long Covid". Patients report a range of symptoms, including extreme tiredness (fatigue), shortness of breath, loss of taste and smell, chest pains and impaired cognition ("brain fog").

During my master's degree in Human-Computer Interaction, I led a team tasked with producing an app to support people suffering from Long Covid. Through research, we found patients mentioning fatigue and brain fog as particularly debilitating. We didn't feel a digital experience would help resolve the remaining symptoms, given their life-threatening nature or unclear origins.

The Solution

Energy envelope theory

Different things contribute to fatigue and can make it last a long time. Low levels of physical activity, a disturbed daily routine, poor sleep patterns, demanding work, caring responsibilities, low mood, anxiety and stress can all make fatigue worse. Our app, LongCo, helps patients manage and reduce these factors by promoting energy conservation behaviours such as pacing, planning and prioritisation.

It was designed based on energy envelope theory which suggests that patients who maintain their expended energy at a level consistent with their available energy will have better health outcomes and quality of life compared to those who over-expend their energy levels. I led the experience design, facilitating efforts to scope, prototype and evaluate LongCo.

High-level goals:
  • - motivate patients to engage in self-management and fatigue-tracking;

  • - reduce fatigue levels and increase energy;

  • - tailor support according to individual differences;

    - and help clinicians remotely monitor patient progress.


Due to ethical concerns around conducting research with vulnerable groups, UCL restricted us from collecting data directly from patients. To overcome this constraint, we used existing resources such as peer-reviewed studies, patient stories, online support groups, and news articles.

This research approach presented several challenges, including rapidly changing information and conflicting medical advice. Some clinicians argue that post-viral fatigue originates from irrational beliefs and avoidance behaviours, recommending cognitive behavioural therapy and graded exercise. Whereas others suggest this is an ineffective and potentially harmful treatment combination. These inconsistencies made it difficult to select a focus for LongCo. After toying around with a few ideas, we decided to base our design on energy envelope theory as it was the most supported treatment method in patient accounts.

User persona image
Up facing arrow icon
One of three personas created during the project. Illustration by Pablo Stanley

Behaviour Design

Using the insights gathered from patient stories and medical journals, we mapped out the key behaviours mentioned by patients and how they relate to each other. This included problem behaviours that contribute to fatigue and desirable behaviours that support adaptive energy conservation (e.g. taking breaks). Through this exercise, we identified "daily reflection and planning" as an activity that would most influence other desired behaviours. To understand the barriers for patients in completing this target behaviour, we undertook a COM-B analysis. Four working hypotheses emerged:

  • patients have limited knowledge about pacing behaviours or recovery times;
  • they lack prompts or cues that encourage energy conservation behaviours;
  • they have unrealistic reflective motivations and social pressures around fast recovery;
  • and low automatic responses to new energy limits.
Behaviour map showing current and potential behaviours
Up facing arrow icon
Behaviour map showing current and potential behaviours


The COM-B model of behaviour is widely used to identify what needs to change for a behaviour change intervention to be effective. It suggests behaviours are produced by influencing one or more of the following: capability, opportunity and motivation.

COM-B analysis diagram
Up facing arrow icon
COM-B analysis for Professional Paula. Target behaviour: recording mood, activity and fatigue levels in app for two mins daily to support self-managed pacing

Design Exploration

We then regrouped to explore how a digital interface might enable users to complete our "daily reflection and planning" target behaviour. The output from this session was a set of high-level user flows that all team members were aligned on. They helped us identify several opportunities for LongCo:

  • creating a personalised experience for each user;
  • including a knowledge hub where users can explore recent articles, research paper summaries and patient success stories;
  • making it easy for users to review goals and self-monitor progress;
  • integration into existing care pathways; clinicians could recommend our app to Long Covid patients;
  • and partnering with research institutions and health services to support their efforts in understanding and treating Long Covid.
Image showing user flow
Up facing arrow icon
Path taken for typical users when setting a goal (top) and completing a daily check-in (bottom)

We tested the experience in four rounds of think-aloud sessions with peers, each time making iterations based on the findings. Our aim was to evaluate usability rather than utility as participants were not representative of our target population.

Image showing LongCo early wireframes
Up facing arrow icon
Low-fidelity wireframes

Final Design

LongCo: Your Path to Living Well With Long Covid

LongCo helps patients manage and overcome fatigue by training them to use energy conservation strategies. Users are encouraged to self-monitor the impact of each strategy on their symptoms and daily goals. Through continued usage, patients learn which behaviours work best for them.


First-time users are presented with a brief summary of the benefits of using LongCo. We had initially included more information about energy envelope theory, but insights from user testing told us the onboarding flow was too long. In our final design, we moved this information into modal pop-ups that appear when users first navigate to each of the three main areas of the app.

High-fidelity design showing LongCo's onboarding process
Up facing arrow icon
Onboarding flow
High-fidelity design showing modal pop up information in LongCo
Up facing arrow icon
Modal pop-ups appear for first-time users

Set Meaningful Goals

The goal-setting flow captures patient goals without requiring them to set detailed action plans. It introduces the most common energy conservation behaviours and prompts users to commit to a daily check-in.

High-fidelity design showing how to set a goal using LongCo
Up facing arrow icon
Goal-setting flow

Track Progress

During the daily check-in, users are encouraged to reflect on the influence of energy conservation behaviours on their goal progression and general condition (fatigue, mood and other symptoms).

High-fidelity design showing how to complete a daily check-in using LongCo
Up facing arrow icon
Check-in flow

Reinforce Desired Behaviours

LongCo reinforces the daily check-in through (1) detailed feedback on behaviour in the trends tab, (2) articles and videos showing how energy conservation strategies have worked for others in learn tab and (3) playful animations and language throughout the app.

High-fidelity design showing the Home, Trends and Learn sections of LongCo
Up facing arrow icon
Check-in complete animation, Trends tab and Learn tab (right)

Measuring Success

As mentioned earlier, UCL restricted us from conducting research with people experiencing Long Covid symptoms. Given the opportunity to continue this project with access to patients, I would first test the hypotheses identified through our COM-B analysis. This could be done by conducting a series of interviews with clinicians and patients. Next, I would trial a lo-fi version of LongCo with a subset of potential users, tracking:

  • Perceived fatigue. Does LongCo achieve its core purpose of helping patients manage and reduce fatigue?
  • Goal progression. How many goals are patients making progress on?
  • Task completion time. How long does it take users to set a goal and complete a daily check-in?
  • Features retention. For example, are users repeatedly using the learn tab? If not, there's a chance their experience of the feature isn't positive.