Effect of the COVID-19 pandemic on procedure volumes in gastroenterology in the Netherlands

The COVID-19 pandemic has challenged health-care systems across the globe. Reallocation of resources and personnel to COVID-19 wards severely affected all aspects of care, including closure of outpatient clinics and cancellation or postponement of procedures.1, 2 We aimed to quantify the effect of the COVID-19 pandemic on nationwide procedure volumes in gastroenterology in the Netherlands during the early and later stages of the pandemic.

We used claims data extracted from the Dutch national health insurers information system (Vektis) to identify all gastroenterological health-care activities performed from Jan 1, 2019, to March 28, 2021, as part of a diagnosis treatment combination plan with a hepatogastroenterologist as the treating physician. The activities were analysed overall and across procedure types and diagnosis groups of interest (appendix pp 2–5). We compared absolute and relative changes for the index week during the COVID-19 pandemic with the reference week in 2019. We defined three timeframes on the basis of the number of weekly COVID-19 admissions:³ the first wave from week 9 through week 22 of 2020; a subsequent recovery phase with relatively few admissions from week 23 through week 35; and a second wave with high ongoing admission rates from week 36 through to the end of the study period (figure ).

Figure.

Procedure volumes during the COVID-19 pandemic.

Change in procedure volumes per week (moving average) during the COVID-19 pandemic versus the reference week in 2019. (A) Overall; (B) by procedure type; (C) by diagnosis. The number of weekly admissions for COVID-19 are also shown in the upper panel (bars, plotted against the secondary axis). VCE=video capsule endoscopy. ERCP=endoscopic retrograde cholangiopancreatography.

We analysed 980 075 procedures. During the peak of the first COVID-19 wave, the number of procedures decreased by 64·1% (figure). There was a deficit of 43 052 procedures at the end of the first wave, corresponding to a 34·5% decrease versus the reference period in 2019 (p=0·004; appendix p 6). During the recovery phase, the total procedure volume returned to reference levels (–1·2% vs 2019; p=0·590; appendix p 6) and remained stable during the second wave (0·5%; p=0·975; appendix p 6). At the end of the study period, a total deficit of 42 882 procedures remained.

Change in procedure volumes varied across procedure types (p<0·001) and diagnosis groups (p=0·018; figure; appendix p 6). In terms of procedure type, the largest absolute decrease was observed for lower endoscopy (–28 628 procedures, –9·8%), whereas the most pronounced relative decrease was observed for non-endoscopic diagnostic procedures (–30·4%). The smallest difference was observed for urgent interventions (–2·5%). In terms of diagnosis groups, the most pronounced decrease was observed for procedures performed for colorectal cancer screening (–18·6%), motility disorders (–13·4%), or chronic abdominal pain and irritable bowel syndrome (–10·3%). Conversely, a smaller decrease was observed in the number of procedures performed for diagnosis or treatment of pancreatic or biliary neoplasia (–0·9%). During the recovery phase, procedure volumes returned to near or more than reference levels in most diagnosis groups, except for colorectal cancer screening (–49·3%). During the second wave, only small decreases in procedure volumes were observed for most diagnosis groups; the major exception was the colorectal cancer screening programme (2342, 7·3%).

This study illustrates the enormous effect of the COVID-19 pandemic on gastroenterology departments in the Netherlands. This effect was mainly accounted for by a large decrease in procedure volumes during the first COVID-19 wave. Although overall procedure volume levels returned to reference levels during subsequent phases of the pandemic, major deficits remained at the end of the study period. The decrease in the number of procedures performed for colorectal cancer screening and surveillance could have important long-term sequelae. Based on the advanced neoplasia detection rate of 35·9% reported in the Dutch colorectal cancer screening programme,⁴ a large number of patients with advanced neoplasia potentially remain undiagnosed and untreated. Furthermore, postponement or cancellation of procedures for surveillance of people at risk for colorectal cancer might increase their risk of advanced stage colorectal cancer.⁵ Concerns about missed diagnoses are furthered by a 42% decline in the number of colorectal cancer diagnoses during the first months of the pandemic.⁶ Attention should be given to the unfavourable long-term effects on colorectal cancer-related morbidity and mortality because of the temporary suspension of colorectal cancer screening and surveillance programmes. Procedure volumes linked to malignancies more likely to present with urgent symptoms such as pancreatic cancer or cholangiocarcinoma were relatively preserved, which has translated to a smaller decline in the number of biliary and pancreatic cancer diagnoses in 2020.⁷

We observed major reductions in the number of procedures performed for diagnosis or treatment of chronic abdominal pain, inflammatory bowel disease, and other non-malignant disorders. The effect of this decrease on morbidity and quality of life of affected patients could also be significant as a result of delayed diagnosis and treatment.8, 9 Nevertheless, some of these procedures could also have been superfluous and might not have needed to be performed.¹⁰

In conclusion, we observed profound reductions in gastroenterological procedure volumes during the first wave of the COVID-19 pandemic. Procedure volumes returned to or increased above reference levels during subsequent phases of the pandemic, which is a testament to the resilience and creativity of gastroenterology departments across the nation. Nevertheless, a significant backlog of postponed or cancelled procedures remained at the end of the study period; the long-term effects on quality of life, morbidity, and mortality requires careful further evaluation.