The impact of COVID-19 pandemic on calls to Samaritans’ telephone helpline in the UK and ROI

Study objectives: The COVID-19 pandemic led to greater reliance on and uptake of remote services, and this study aimed to ascertain whether the pandemic and restrictions had any impact on behavior of callers to Samaritans, a national crisis helpline.
Methods: Duration of calls made to Samaritans over 4 four-week periods within 2020 were analysed; a 4 week period of calls before confirmation of the first known case of COVID-19 (Pre-COVID-19; 03-Feb-2020 - 01-Mar-2020), a 4-week period of calls following the initial nation-wide lockdown restrictions (Active COVID-19; 30-Mar-2020 - 26-Apr-2020), a 4-week period of calls collected 2 months after lockdown (01-Jun-2020 - 28-Jun-2020) and a 4-week period of calls collected 4 months after lockdown (3-Aug-2020 - 30-Aug-2020). K-means clustering was used to determine the types of callers within the data based on call behavior; subsequent analysis looked at changes in the behavior of these caller types caused by the pandemic/lockdown restrictions at each period.
Results: Fewer calls up to 5 minutes in duration (perceived as “checking-in” calls) were made from the Pre-COVID-19 period to the first Active COVID-19 period, likewise, more calls that are greater than 30 minutes (perceived as an “emotional support” session) were made to the service. Distribution of call duration over the latter two Active COVID-19 periods begins to revert back to a Pre-COVID-19 norm. Clustering revealed 5 distinct caller types within the data; 2 out of the 5 caller types (‘high frequency’ and ‘typical’) appear to be significantly impacted during the data collection period, in relation to their durations of calls (in general and over a 24-hour day).
Conclusion: Changes in caller behavior point to the impact of removing existing face-to-face mental health supports during the pandemic and increasing distress amongst the most “at risk” callers to the service, but not across all caller archetypes. Telephony data can be used to measure the effect of an external event on society in real-time; in this case, the impact of COVID-19 on a national crisis helpline service.