Impact of COVID-19 on outpatient appointments in children and young people in England: an observational study

Objectives To describe the impact of the COVID-19 pandemic on outpatient appointments for children and young people. Setting All National Health Service (public) hospitals in England. Participants All people in England aged <25 years. Outcome measures Outpatient department attendance numbers, rates and modes (face to face vs telephone) by age group, sex and socioeconomic deprivation. Results Compared with the average for January 2017 to December 2019, there was a 3.8 million appointment shortfall (23.5%) for the under-25 population in England between March 2020 and February 2021, despite a total rise in phone appointments of 2.6 million during that time. This was true for each age group, sex and deprivation fifth, but there were smaller decreases in face to face and total appointments for babies under 1 year. For all ages combined, around one in six first and one in four follow-up appointments were by phone in the most recent period. The proportion of appointments attended was high, at over 95% for telephone and over 90% for face-to-face appointments for all ages. Conclusions COVID-19 led to a dramatic fall in total outpatient appointments and a large rise in the proportion of those appointments conducted by telephone. The impact that this has had on patient outcomes is still unknown. The differential impact of COVID-19 on outpatient activity in different sociodemographic groups may also inform design of paediatric outpatient services in the post-COVID period.


GENERAL COMMENTS
This is a well written paper on an analysis of a national (England) hospital administrative database & Hospital Episodes Statistics to describe trends in outpatient appointment (numbers, rates and modes). It provides a useful addition to the literature. If available, it will be useful to describe attendances in community vs Hospital clinics and discuss differences.

Amy Sweeny
Gold Coast Hospital and Health Service, Emergency REVIEW RETURNED 08-Feb-2022

GENERAL COMMENTS
Thank you for giving me the opportunity to review this manuscript. The effect of COVID-19 and its mitigation strategies on health care activity is an important topic to study. The COVID-19 pandemic has inadvertently showed us how we can better deliver a substantial proportion of our health care. This may be particularly advantageous for certain age groups and certain reasons for seeking care, due to the convenience of telehealth rather than in-person appointments. Strengths: Population study using NHS dataset. Reasonable approach not to do any statistical testing due to the size of the dataset Concerns: Hospital cancellations were excludedthis may be an issue, and we need to see some data on what proportion of all appointments this represented.
Appt type face to face, telephone and not specified. How many appointments were not specified? Did this vary by age group? This information needs to be provided. They stated that they did not do a time series analysis due to lack of a trend. Instead they used a three-year average of historic data for 2017-2019 This is somewhat unusual, and some evidence of a lack of a trend over the 3 years would be appropriate to include (e.g. in an appendix).
Minor points: Please clarify if appointment data in Table 1 and Figure 1 is scheduled appointments only, not appointments attended. Please include this information in the title. I am unfamiliar with the health system in the UK. Is a decrease in appointments due to the inability to staff them on behalf of the health system, or a decrease in demand on behalf of the public? The data suggest that appointments (I assume scheduled appointments) dropped considerably, or by nearly one-quarter. The implications of this are that the health system reduced care offered, is this correct? Some more information here would be helpful for the international audience. On page 7 line 14, please explain what you mean by absolute percentage.
Reconciliation of Table 1 with Table 2 is difficult. It seems that phone appointments might be a larger proportion of all appointments based on the numbers in Table 1. It would be helpful to see the actual numbers rather than just the amount decreased or increased. The proportion of first phone appointments for under 1s remains quite low (2.8 -4.6% of all appointments). The change from prepandemic appears to be overstated. But the authors then state that face to face appointments for the under 1s have been "relatively preserved" (line 41, p9), although the tables suggest that they had dropped by 28% (Table 1), and decreased (Table 2). Some clarification of their conclusions, and/or enhancing the tables to include raw numbers not just number decreased (previously suggested) would be helpful.

VERSION 1 -AUTHOR RESPONSE
3 Reviewer 1: If available, it will be useful to describe attendances in community vs Hospital clinics and discuss differences.
Thank you for this suggestion and we agree that this is an important issue. Unfortunately, this data is not available in the dataset. In our strengths and limitations section we mention that future work would benefit from linkage with primary care records to look at community attendances.
Reviewer 2: Hospital cancellations were excludedthis may be an issue, and we need to see some data on what proportion of all appointments this represented.
Thank you for this important point. We have now analysed the proportion of cancellations and included this in the text. The percentage of cancellations ranged from 7.7% in the under-1year-olds to 10.4% in the 1-4-year-olds in 2019-2020. However, are unable to analyse what proportion of these were rebooked, although in our experience it is high.
Reviewer 2: Appt type face to face, telephone and not specified. How many appointments were not specified? Did this vary by age group? This information needs to be provided.
For the whole period, the percentage of not specified appointments ranged from 0.2% in the under-1-year-olds to 0.6% in the 15-19-yearolds. We have now included this information in the text.
Reviewer 2: They stated that they did not do a time series analysis due to lack of a trend. Instead they used a three-year average of historic data for 2017-2019 This is somewhat unusual, and some evidence of a lack of a trend over the 3 years would be appropriate to include (e.g. in an appendix). Figure S1 shows trends for each age since 2018. We have performed some reanalysis and the log RR of the monthly linear trend is just -0.00006 for all ages combined (RR of 0.99994) and varied little by age. We have added this information to Figure S1.
Reviewer 2: Please clarify if appointment data in Table 1 and Figure 1 is scheduled appointments only, not appointments attended. Please include this information in the title.
It is scheduled appointments. We have amended the Table 1 and Figure S1 titles to reflect this.
Reviewer 2: I am unfamiliar with the health system in the UK. Is a decrease in appointments due to the inability to staff them on behalf of the health system, or a decrease in demand on behalf of the public? The data suggest that appointments (I assume scheduled appointments) dropped considerably, or by nearly one-quarter. The implications of this are that the health system reduced care offered, is this correct? Some more information here would be helpful for the international audience.
Thank you for this important suggestion. We have added some more information about the different reasons for the decrease in activity into the 'Findings compared with previous studies' section of the paper and included two new references that demonstrate similar changes in primary and emergency care.