BACKGROUND: Most people with COVID-19 self-manage at home. However, the condition can deteriorate quickly and some may develop serious hypoxia with relatively few symptoms. Early identification of deterioration allows effective management with oxygen and steroids. Telemonitoring of symptoms and physiological signs may facilitate this.
OBJECTIVE: To design, implement and evaluate a telemonitoring system for people with COVID-19 self-managing at home considered at significant risk of deterioration.
METHODS: A multi-disciplinary team developed a telemonitoring protocol using a commercial platform to record symptoms, pulse oximetry and temperature. If symptoms or physiological measures breached targets, patients were alerted asking them to phone an ambulance (red) or for advice (amber). Patients attending COVID assessment centres, considered fit for discharge but at risk of deterioration, were shown how to use a pulse-oximeter and the monitoring system which they were to use twice daily for two weeks. Patients could interact by app, SMS or touch-tone phone. Written guidance on alerts was also provided. Following consent, patient data on telemonitoring usage and alerts were linked to data on service resource use. Subsequently, patients who had both used and not used the telemonitoring service, including those who had not followed advice to seek help, agreed to brief telephone interviews to explore their views on and how they had interacted with the telemonitoring system. Interviews were recorded and analysed thematically. Professionals involved in the implementation were sent an online questionnaire asking them about their perceptions of the service.
RESULTS: We investigated the first 116 patients who used the service. Of these 71 (61%) submitted data, the remainder chose to self-monitor without electronic support. Of the 71 patients who submitted data, 35 received 151 alerts during their two-week observation. Sixty-seven 'red' alerts were for oxygen saturation (Sa02) levels ≤93% and 15 because they recorded severe breathlessness. Nineteen were admitted to hospital (average stay 3.4 days). Of the 45 who used written guidance alone, eight were admitted to hospital (average stay 5.3 days) and one died. Some patients who were advised to seek help did not do so, some because parameters improved on re-testing, others because they felt no worse than before. All patients found self-monitoring reassuring. Most professionals who used the system (n=11) found it easy to use and useful. Five professionals considered the system 'very safe', three thought it 'could be safer', and three wished more experience before deciding. Two felt SaO2 trigger thresholds were too high.
CONCLUSIONS: Supported self-monitoring of patients with COVID-19 at home is reassuring to patients acceptable to clinicians and can detect important signs of deterioration. Worryingly, some patients, because they felt well, occasionally ignored important signs of deterioration. It is important therefore to emphasise the importance of the early investigation and treatment of asymptomatic hypoxia at the time when patients are initiated and in the warning messages that are sent to patients.
CLINICALTRIAL: Not applicable.