Changes in colorectal cancer treatment during the COVID-19 pandemic in Japan: Interrupted time-series analysis using the National Database of Japan

Abstract

Background

The coronavirus disease 2019 (COVID-19) pandemic forced us to accept changes in our usual diagnostic procedures and treatments for colorectal cancer. This study aimed to determine the impact of the pandemic on colorectal cancer treatment in Japan.

Methods

The number of colorectal surgeries, stoma constructions, stent placements or long tube insertions, and neoadjuvant chemoradiotherapies were determined each month using sampling datasets from the National Database of Health Insurance Claims and Specific Health Checkups of Japan. The observation periods before and during the pandemic were January 2015 to January 2020 and April 2020 to January 2021, respectively. An interrupted time-series analysis was used to estimate the changes in the number of procedures during the pandemic.

Results

The number of endoscopic surgeries for colon cancer significantly decreased in April and July 2020 and for rectal cancer in April 2020. Additionally, the number of laparoscopic and open surgeries for colon cancer significantly decreased in July 2020 and October 2020, respectively. The number of stoma constructions and stent placements or long tube insertions did not increase during the observation period. Neoadjuvant chemoradiotherapy for rectal cancer significantly increased in April 2020 but levels returned shortly thereafter. These results suggest that the recommendations to overcome the pandemic proposed by expert committees, including the replacement of laparoscopic surgery with open surgery, stoma construction to avoid anastomotic leak, and replacement of surgery on the ileus with stent placement, were not widely implemented in Japan. However, as an exception, neoadjuvant chemoradiotherapy for rectal cancer was performed as an alternative treatment to delay surgery in small quantities.

Conclusion

A declining number of surgeries raises concerns about cancer stage progression; however, we found no evidence to suggest cancer progression from the trajectory of the number of stoma constructions and stent placements. In Japan, even during the pandemic, conventional treatments were performed.

Graphical Abstract

1. Introduction

The coronavirus disease 2019 (COVID-19) pandemic has placed unprecedented pressure on healthcare providers responsible for managing cancers, including colorectal cancer. To prevent infection and accommodate limited medical resources resulting from the pandemic, they had to adopt changes in their usual diagnostic procedures and standard treatments. Cancer screening was suspended at an early stage of the pandemic in various countries [1], [2], [3], [4], [5] and the number of cancer diagnoses declined [6], [7]. Additionally, several expert committees issued guidelines for the management of cancer during the pandemic to help healthcare providers make difficult decisions under unprecedented circumstances [8], [9], [10], [11], [12]. These guidelines recommended postponing nonurgent gastrointestinal endoscopy and nonfatal or nonurgent cancer surgery [9], [11], [12]. A preference for open surgery over laparoscopic or robotic surgery was suggested because of concerns about increased risk of aerosol spread of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) during the laparoscopic procedure [10]. To delay surgery until the peak of COVID-19 has passed and resources have returned, alternative treatments, including neoadjuvant chemoradiotherapy, were proposed [9], [10], [12]. Lower-risk strategies, including stent placement for patients with obstructive colorectal cancer rather than surgery and Hartmann’s procedure rather than resection with primary anastomosis, were recommended [9], [10], [12]. Thus, colorectal cancer treatment during the pandemic changed in various countries [5], [13], [14], [15], [16]; however, these changes vary depending on differences in the health insurance system, extent of the spread of COVID-19, extent of movement restriction, and standard treatment modalities before the pandemic in each country.

In Japan, the first person infected with SARS-CoV-2 was identified on January 15, 2020 and three emergency declarations were issued. In the early stage of the pandemic, the extent of the spread of COVID-19 was relatively low, and no lockdown was implemented even under the emergency declarations [17], [18], [19]. The prefectural governments requested residents and institutions to take action to prevent infection without enforcing action or penalties on violators. In terms of colorectal cancer treatment, the standard treatment for patients with locally advanced rectal cancer differs between Japan and other countries; resection is recommended as the first treatment choice in Japan, while neoadjuvant therapy is recommended in the United States [20], European countries [21], [22], [23], and China [24]. Several studies have reported the impact of the COVID-19 pandemic on colorectal cancer diagnosis and treatment in Japan. Similar to other countries, cancer screening was suspended at an early stage of the pandemic, and the number of colorectal cancer screenings conducted by local governments has decreased [25]. Additionally, the number of procedures commonly used for colorectal cancer diagnosis in clinics and hospitals, including colonoscopy and colorectal biopsy, decreased during the pandemic [26]. Consequently, the number of colorectal cancer diagnoses [27], [28], [29] and surgeries [30], [31] declined during the pandemic. However, changes in the numbers according to the type of surgery, including endoscopic, laparoscopic, or open surgery, have not been reported. Additionally, several studies have suggested the progression of colorectal cancer stages during the pandemic by comparing the stages at diagnosis, number of emergency admissions, number of obstructive colorectal cancers, and number of colorectal stent placements before and during the COVID-19 pandemic [32], [33], [34]. However, these were all single-institution studies. Therefore, in determining the impact of the pandemic on Japan, generalizability is limited.

This study aimed to determine changes, if any, in the treatment of colorectal cancer during the COVID-19 pandemic in Japan using the National Database of Health Insurance Claims and Specific Health Checkups of Japan (NDB), in which insurance claims data were accumulated comprehensively. This study aimed to clarify the impact of the COVID-19 pandemic on colorectal cancer treatment and provide suggestions about cancer stage progression in Japan.

2. Methods

2.1. Data sources

This was a repeated cross-sectional study using sampling datasets from the NDB. The NDB was established in 2008 by the Ministry of Health, Labour, and Welfare of Japan (MHLW) based on the Act on Assurance of Medical Care for Elderly People. As of March 31, 2020, 18,768 million insurance claims data have accumulated. For example, in July 2022, 59,509,947 medical insurance claims were registered in the database, corresponding to 98.5% of all claims in Japan. The NDB data are provided to researchers for utilization in research after reviewing the data management plan and research objectives. Depending on the data extraction method, there are three types of data to provide. Of those, we used the sampling datasets, whose figures are follows; 1. Sampling datasets were created independently for each month. Therefore, it was impossible to link each month; 2. Target months were 4 months per year, January, April, July, and October. 3. Data were extracted randomly from the NDB at an extraction rate of 1% for outpatient claims and 10% for inpatient and diagnosis procedure combination (DPC) claims. Pharmacy claims linked to outpatient claims by hash values to identify individuals created by insurer number, insured number, date of birth, and sex were also extracted. 4. No personally identifiable information, including names or addresses, was included. For this study, sampling datasets from January 2015 to January 2021 (25 months) were provided by the MHLW, including 20,452,831 medical outpatient claims, 3115,714 medical inpatient claims, and 2507,790 DPC claims.

2.2. Outcomes and extraction procedures

Insurance claims for the target treatments were extracted from the sampling datasets. The extraction procedure is described in Supplementary File 1. Surgeries were divided into six types: endoscopic, laparoscopic, and open surgeries for colon or rectal cancer. For surgeries for colon cancer, insurance claims with the disease name code of colon cancer corresponding to the ICD–10 codes of C18.0 and C18.2–9 were targeted. In addition, for surgeries for rectal cancer, patients with rectal cancer corresponding to the ICD-10 codes of C19 and C20 were targeted. The disease name codes are listed in Supplementary Table 1.

As an indicator of obstructive colorectal cancer, the claims with the medical practice codes of “stent placement” or “long tube insertion for ileus” were extracted, where claims with the disease name code of colon and rectal cancer were targeted. Stoma construction was divided into three types: stoma construction with rectal amputation, rectal resection, and without resection and amputation. For the former two, claims with a disease name code for rectal cancer were targeted; for the latter, claims with a disease name code for colon or rectal cancer were targeted. As an additional payment charge covering stoma construction along with colon resection was newly established in April 2020, this was excluded from the evaluation.

2.3. Creating time series data

We created time-series data (Supplementary File 2) with 25 points for each outcome using the following procedure: First, we counted the number of claims for each insurance claim type (outpatient, inpatient, and DPC) for each month. Then, the counted number was multiplied by 100 for outpatient claims and by 10 for inpatient and DPC claims (multiplied by the reciprocal of the extraction rate). Finally, for endoscopic surgery for colon or rectal cancer, the total number of outpatient, inpatient, and DPC claims were determined. For other surgeries and practices, the total number of inpatient and DPC claims was determined because the number of extracted outpatient claims was zero or very small.

2.4. Statistical analysis

We performed interrupted time-series analysis using the seasonal autoregressive integrated moving average model (SARIMA), as described in a previous study [26]. The analytical procedure was based on the method described by Schaffer et al. [35]. All statistical analyses were performed using R version 4.2.0, an open-source software package. First, we determined the SARIMA components, (p, d, q)× (P, D, Q)s, with the lowest Akaike’s information criterion, using auto.arima function in the forecast package for R. Seasonality (s) was determined to be 4 because our time-series data were by quarter. In this step, all 25-point time-series data were used and step-change variables during the pandemic were included in the model as external regressors. The step-change variables take the value of one for each month during the pandemic and zero otherwise. The estimates of the step-change variables represent the changes in numbers for each month during the pandemic. Second, the number during the pandemic was predicted assuming that the pandemic was absent (counterfactual number) using only pre-pandemic points (21 points). In this step, the SARIMA components are applied to those determined in the first step. The rate of change was calculated by dividing the estimated change by the counterfactual number. For stoma construction along with rectal resection, analysis was performed using time-series data after July 2018 because the additional payment charged when a stoma was created along with rectal resection was newly established in April 2018, and the numbers seemed to stabilize after July 2018.

3. Results

The estimated changes in the number of colon and rectal cancer surgeries are shown in Table 1, and the observed and counterfactual numbers are shown in Fig. 1. The number of endoscopic surgeries for colon significantly decreased in April 2020 (−1824; 95% confidence interval [CI]: −3239, −409) and July 2020 (−1426; 95% CI: −2842, −10) and that for rectal cancer in April 2020 (−1015; 95% CI: −1962, −68). The number of laparoscopic and open surgeries for colon cancer significantly decreased in July 2020 (−703; 95% CI: −1246, −161) and October 2020 (−354; 95% CI: −693, −16), respectively; those for rectal cancer did not change. No increase in the three types of stoma construction, those with rectal amputation, those with rectal resection, and those without amputation and resection, was observed ( Table 2); the observed and counterfactual numbers with 95% CIs are shown in Fig. 2a, b, and c. The number of stent placements or long tube insertions for the ileus significantly decreased throughout the observation period (Table 2); the observed and counterfactual numbers are shown in Fig. 2d. The number of neoadjuvant chemoradiotherapies for rectal cancer significantly increased in April 2020 (71; 95% CI: 37, 104). (Table 2, Fig. 2e).

Table 1.

Estimated changes in number and rate of colon and rectal cancer surgeries.

			Estimated changes in number			Estimated changes in rate (%)
Site	Surgery type	Time	Numbers	95% CI		Rate (%)	95% CI		p-value
Colon	Endoscopic	2020/04	-1824	-3239	-409	-19	-34	-4	0.012
		2020/07	-1426	-2842	-10	-15	-29	0	0.048
		2020/10	99	-1576	1775	1	-15	17	0.908
		2021/01	-302	-2042	1439	-3	-21	15	0.734
	Laparoscopic	2020/04	-398	-837	41	-10	-20	1	0.075
		2020/07	-703	-1246	-161	-17	-30	-4	0.011
		2020/10	-397	-939	144	-10	-23	4	0.151
		2021/01	-14	-576	548	0	-14	14	0.962
	Open	2020/04	3	-303	310	0	-16	16	0.984
		2020/07	-246	-565	73	-13	-31	4	0.130
		2020/10	-354	-693	-16	-18	-34	-1	0.040
		2021/01	-76	-446	294	-5	-28	18	0.687
Rectum	Endoscopic	2020/04	-1015	-1962	-68	-26	-51	-2	0.036
		2020/07	-504	-1458	450	-12	-35	11	0.300
		2020/10	381	-619	1381	11	-18	39	0.455
		2021/01	-803	-1956	350	-19	-47	8	0.172
	Laparoscopic	2020/04	270	-58	598	14	-3	30	0.107
		2020/07	250	-214	714	13	-11	36	0.291
		2020/10	50	-519	619	3	-26	31	0.863
		2021/01	-20	-677	637	-1	-34	32	0.952
	Open	2020/04	-38	-233	157	-7	-42	29	0.703
		2020/07	-40	-327	247	-8	-63	47	0.784
		2020/10	60	-359	480	13	-80	107	0.778
		2021/01	29	-395	453	5	-75	86	0.894

CI: confidence interval

Fig. 1.

Trajectories of the number of surgeries for colon or rectal cancer, a indicates the number of endoscopic surgeries for colon cancer, b indicates the number of laparoscopic surgeries for colon cancer, c indicates the number of open surgeries for colon cancer, d indicates the number of endoscopic surgeries for rectal cancer, e indicates the number of laparoscopic surgeries for rectal cancer, f indicates the number of open surgeries for rectal cancer, The number of endoscopic surgeries was calculated using the medical outpatient, inpatient, and diagnosis procedure combination insurance claims data, The numbers of laparoscopic and open surgeries were calculated using medical inpatient and diagnosis procedure combination insurance claims, The observed numbers during all observation periods (solid line) and the expected counterfactual numbers with 95% confidence intervals (CIs) during the COVID-19 pandemic (dashed and dotted lines) were drawn.

The data were represented by a quarterly series: January, April, July, and October.

Table 2.

Estimated changes in number and rate of stoma constructions, stent placements or long tube insertions, and neoadjuvant chemoradiotherapies.

			Estimated changes in number			Estimated changes in rate (%)
Sites	Practices	Time	Numbers	95% CI		Rate	95% CI		p-value
Rectum	Stoma construction with rectal amputation	2020/04	68	-51	186	18	-14	50	0.265
		2020/07	-62	-181	56	-17	-49	15	0.304
		2020/10	-12	-131	106	-3	-35	29	0.838
		2021/01	-52	-171	66	-14	-46	18	0.388
Rectum	Stoma construction with rectal resection	2020/04	-33	-118	52	-5	-20	9	0.449
		2020/07	17	-68	102	3	-11	17	0.693
		2020/10	-93	-178	-8	-15	-30	-1	0.032
		2021/01	57	-28	142	9	-5	24	0.188
Colon and rectum	Stoma construction without amputation and resection	2020/04	-122	-367	122	-19	-56	18	0.326
		2020/07	136	-62	333	21	-10	52	0.179
		2020/10	75	-126	275	12	-21	45	0.464
		2021/01	-73	-274	127	-11	-42	20	0.473
Colon and rectum	Stent placement or long tube insertion	2020/04	-209	-393	-25	-12	-23	-1	0.026
		2020/07	-286	-491	-81	-15	-25	-4	0.006
		2020/10	-211	-416	-7	-10	-20	0	0.043
		2021/01	-404	-600	-207	-22	-33	-11	< 0.001
Rectum	Neoadjuvant chemoradiotherapy	2020/04	71	37	104	120	63	177	< 0.001
		2020/07	31	-3	64	52	-4	109	0.070
		2020/10	-29	-63	4	-49	-106	8	0.089
		2021/01	-19	-53	14	-32	-89	25	0.265

CI: confidence interval

Fig. 2.

Trajectories of the number of stoma constructions, stent placements of long tube insertions, and neoadjuvant chemoradiotherapies, a indicates the number of stoma constructions along with rectal amputation, b indicates the number of stoma constructions along with rectal resection, c indicates the number of stoma constructions without amputation and resection, d indicates the number of stent placements or long tube insertions for ileus, e indicates the number of neoadjuvant chemoradiotherapy for rectal cancer, The numbers were calculated using medical inpatient and diagnosis procedure combination insurance claims, The observed numbers during all observation periods (solid line) and the expected counterfactual numbers with 95% confidence intervals (CIs) during the COVID-19 pandemic (dashed and dotted lines) were drawn, For stoma construction along with rectal resection, analysis was performed using time-series data after July 2018 because the additional payment charged when a stoma was created along with rectal resection was newly established in April 2018, and the numbers seemed to stabilize after July 2018, The data were represented by a quarterly series: January, April, July, and October.

4. Discussion

This study revealed the changes in the number of surgeries and practices for colorectal cancer treatment before and during the COVID-19 pandemic using sampling datasets from the NDB in Japan.

The number of endoscopic surgeries performed for colon cancer significantly decreased in April and July 2020 and for rectal cancer in April 2020. Endoscopic surgery is performed for early stage colorectal cancers, including intramucosal carcinoma or mildly invasive carcinoma of the submucosa. At the beginning of the pandemic, colorectal cancer screening was suspended in various countries [1], [2], [3], [4], [5] and the number of cancer diagnoses decreased [6], [7]. Japan is no exception. According to the analysis of hospital-based cancer registries, the number of colorectal cancer diagnoses, especially cases detected by screening, decreased in 2020 compared with the average number in the past four years [27]. Moreover, the decline in the diagnosis of early stage cancer during the pandemic was more pronounced than that of advanced cancers [27], [28], [29], [32], [36]. Our results show that endoscopic surgery, which is performed in patients with low-stage colorectal cancer, decreased during the pandemic, consistent with the findings of previous studies. Although laparoscopic and open surgeries for colon cancer decreased only in July and October 2020, respectively, the decline in these surgeries lags behind that of endoscopic surgeries. Endoscopic surgery is sometimes performed immediately after a colonoscopy when abnormalities are detected. Contrastingly, laparoscopic and open surgeries are performed only after the diagnosis is confirmed by endoscopy and biopsy. Therefore, we assumed that the timing of the decrease in laparoscopic and open surgeries is somewhat later than that of endoscopic surgery. Additionally, the use of laparoscopic and open surgery decreased for colon cancer but not for rectal cancer. This cannot be explained by our data, but it could be because rectal cancer is more symptomatic than colon cancer. Regardless of the reason, no increase was observed to compensate for the decrease in the number of surgeries performed during the observational period. As delayed surgery for colon cancer, even those of 30–40 days, is associated with lower survival and worse outcomes [37], [38], a worse prognosis in the future is a concern.

According to the guidelines for the management of cancer during the COVID-19 pandemic era issued by The Association of Coloproctology of Great Britain and Ireland (ACPGBI), open surgery should be preferred over laparoscopic procedures unless there are specific circumstances where the patient will benefit from the use of a laparoscopic approach, because an increased risk of aerosol spread of SARS-CoV-2 during laparoscopic procedures has been reported [10]. In a systematic review and meta-analysis, minimally invasive approaches were more commonly used in the pre-pandemic group than in the pandemic group, although the difference was not statistically significant [14]. A registry-based study in England [5] and single-institution studies in Japan [34] and China [16] reported a higher proportion of open surgeries during the pandemic than in the pre-pandemic period. Unlike ACPGBI, the Japan Surgery Society stated that there is no evidence to support open surgery over minimally invasive surgery in their guidelines [9]. In this study, the numbers of laparoscopic and open surgeries for colon cancer decreased during the pandemic, whereas those for rectal cancer did not change. These results suggest that laparoscopic surgery has not replaced open surgery in Japan during the pandemic. The recommendations from the Japan Surgery Society might have some influence on the minimal change [9].

This study revealed that the number of stomas constructed did not increase during the COVID-19 pandemic. A registry-based study in England [5], Australia, and New Zealand [39] reported that the number of patients who underwent stoma increased during the pandemic, whereas a meta-analysis reported no change [14]. There are three main reasons for the stoma construction. First, a stoma is constructed permanently along with rectal amputation when there is a malignant tumor near the anus that cannot be preserved. Although anus-sparing surgery has recently become possible even in cases where rectal cancer develops in the lower rectum, rectal amputation is performed and a stoma is required for advanced rectal cancer with invasion of the anal sphincter. In the present study, the number of stoma constructions along with rectal amputation did not change before and during the pandemic, suggesting no increase in advanced rectal cancer during the observation period. Second, a stoma is temporarily constructed to avoid the passage of excrement through the surgical anastomosis of the gastrointestinal tract. The guidelines for colorectal cancer management from the ACPGBI [10] and Italian Society of Colorectal Surgery [40] recommend stoma construction to reduce the risk of anastomotic leaks, including Hartmann’s procedure, instead of resection with primary anastomosis. A study in England reported that the number of stoma constructions increased during the pandemic and suggested that the stoma had been intended as a temporary measure until it could be reversed to restore gastrointestinal continuity in many of the cases [5]. Contrastingly, our study noted no increase in any type of stoma construction, suggesting that stoma formation was not performed as a measure to avoid anastomotic leaks during the pandemic in Japan. Third, a stoma is created to decompress the bowel in patients with intestinal obstructions or strictures due to colorectal cancer. Another approach for intestinal obstruction is decompression by stent placement or long tube insertion. A registry-based study in the Netherlands reported that the number of patients with colon cancer presenting with the ileus increased during the pandemic [41]. Similarly, single-institution studies in Japan reported an increase in the number of obstructive colorectal cancers and colorectal stent placements [33], [34]. Mizuno et al. suggested that delays in the detection and surgery of colorectal cancer during the pandemic may have promoted progression and resulted in an increase in the number of obstructive colorectal cancers [33]. However, this interpretation remains questionable because both studies in Japan were single-institution studies. The hospitals where these studies were performed provided cancer care as usual during the pandemic. Contrastingly, neighboring medical facilities may have struggled to provide cancer treatment because of the burden of providing COVID-19 care. Therefore, advanced cancer patients who would have been treated at other medical institutions if the pandemic did not occur, may have been referred to these hospitals during the pandemic. An increase in the number of obstructive colorectal cancers and stent placements may be a feature found only in such hospitals. To determine the overall impact of the pandemic on Japan, it is necessary to use generalizable data. We utilized the NDB, which comprehensively accumulated insurance claims data in Japan, and found that the number of stoma constructions and stent placements or long tube insertions for the ileus did not increase during the pandemic. Therefore, intestinal obstruction was not thought to have increased in Japan as a whole during the pandemic, and no evidence supporting cancer progression was found during our observational period.

Based on the trajectories of the number of stoma constructions and stent placements or long tube insertions, we discuss adherence to the guidelines for managing obstructive colorectal cancer during the pandemic. Guidelines from expert committees including the Japan Surgical Society recommended a stent placement for the patients with obstructive colorectal cancer as a bridge to surgery [9], [10], [12], [40]. Our results indicate that stent placement or long tube insertion for the ileus decreased rather than increased. This result suggests that the recommendations in the guidelines have not been actively adopted in Japan. However, we cannot explain why stent placement or long tube insertion for the ileus had consistently declined during the pandemic. It is unlikely that the incidence of intestinal obstruction decreased owing to the COVID-19 pandemic. Patients with intestinal strictures may have been treated with fasting and intravenous fluids until their conditions improved instead of performing surgery or stenting.

The use of neoadjuvant chemoradiotherapy for rectal cancer has increased significantly in April 2020. Guidelines for colorectal cancer management during the pandemic recommend neoadjuvant chemoradiotherapy or radiotherapy for patients with rectal cancer as alternative treatments to delay surgery [10], [12], [40]. An increase in neoadjuvant therapy during the pandemic was reported in a meta-analysis [14] and a registry-based study in England [5]. However, unlike the United States [20], European countries [21], [22], [23], and China [24], neoadjuvant therapy is not the standard first choice treatment for locally advanced rectal cancer in Japan. Contrary to our expectations, the number of neoadjuvant chemoradiotherapies for rectal cancer significantly and temporarily increased during an early stage of the pandemic. However, the numbers during the observational period were only approximately 50–100 per month, and the estimated change number was 71, with a 95% CI of 37–104. Neoadjuvant chemoradiotherapy for rectal cancer was performed on a small scale before the pandemic in Japan, and may have been temporarily used as an alternative treatment to delay surgery during the pandemic. If this is true, the number of laparoscopic and open surgeries should decrease to reflect the increase of neoadjuvant chemoradiotherapy, however, such a trajectory was not seen. The numbers of laparoscopic and open surgeries were 20 and 5 times larger than that of neoadjuvant chemoradiotherapies. Therefore, the reduced surgery with neoadjuvant chemoradiotherapy may be masked by random error in the number of surgeries.

This study had some limitations. First, the disease name code was used to extract the target claims. However, this is only for billing purposes and not for diagnostic or therapeutic purposes. Therefore, there was a fluctuation in the code. Second, the NDB sampling dataset was quarterly. Therefore, monthly changes were not evaluated. In addition, changes that occurred in months that were not included in the sampling datasets could not be captured. Third, the observation period during the pandemic may have been too short to assess the impact of the pandemic as a whole. However, we have used available sampling datasets from the NDB wherever possible. Fourth, although we suspected that neoadjuvant chemoradiotherapy may have been used as an alternative treatment to delay surgery, we could not confirm whether surgery for rectal cancer was performed after chemoradiotherapy, because sampling datasets cannot be linked at an individual level.

The greatest strength of this study lies in the use of the NDB, in which insurance claims data in Japan have been comprehensively accumulated. To our knowledge, this is the first study to investigate the impact of the COVID-19 pandemic on colorectal cancer treatment using data from the NDB, which has high generalizability. The second strength is the use of interrupted time-series analysis with the SARIMA model. Previous studies have simply compared the pre- and during pandemics, or in 2020 and 2019. Therefore, pre-existing underlying trends, autocorrelation, moving average, and seasonality were not considered in previous studies, whereas our study controlled for these by an interrupted time-series analysis with the SARIMA model. This analysis is considered one of the best designs for establishing causality when randomized controlled trials are feasible [35].

5. Conclusion

The number of endoscopic surgeries for colon and rectal cancer and laparoscopic and open surgeries for colon cancer decreased temporarily and the number did not increase to compensate for the decrease thereafter. Therefore, worsening prognosis due to untreated cancer is a concern. However, our data showed no increase in the number of stomas and stent placements or ileus tube insertions, indicating no evidence of a progression of cancer stages during our observational period. Several expert committees have suggested changes in cancer management to protect patients and healthcare providers from SARS-CoV-2 in the guidelines; however, they entail deviations from the accepted standards of care. In Japan, even during the pandemic, the usual treatment that would occur normally was maintained. However, as an exception, neoadjuvant chemoradiotherapy for rectal cancer was performed as an alternative treatment to delay surgery at an early stage of the pandemic. The impact of the pandemic varies among countries, therefore, in future, it is necessary to integrate data from various countries to understand global impacts. This study could be used as a source of this data and could contribute to understanding the global impact of the pandemic on colorectal cancer treatment, not just in Japan.

Funding statement

This study did not receive any specific grants from funding agencies in the public, commercial, or non-profit sectors.

Ethical statement

The NDB sampling dataset did not contain personally identifiable information. Therefore, informed consent was not required. The Research Ethics Committees of the Chiba Foundation for Health Promotion and Disease Prevention approved this study (approval number R3-4), which was conducted in accordance with the principles of the Declaration of Helsinki and the Ethical Guidelines for Medical and Biological Research Involving Human Subjects.

CRediT authorship contribution statement

All the authors contributed to the conception and design of the study. Offer to provide the NDB sampling dataset was performed by Misuzu Fujita and Akira Hata. The data extraction conditions were determined by Misuzu Fujita, Kiminori Suzuki, Tokuzo Kasai, Hideyuki Hashimoto, Kazuya Yamaguchi, Daisuke Sato, Takehiko Fujisawa, and Akira Hata. Data analysis plan was determined by Misuzu Fujita and Kengo Nagashima, and data analysis was performed by Misuzu Fujita. The first draft of the manuscript was written by Misuzu Fujita, and all the authors commented on the previous versions of the manuscript. All authors have read and approved the final manuscript.

Declaration of Competing Interest

The authors have no competing interests to declare.

Appendix A. Supplementary material

Supplementary material

.

Supplementary material

.

Supplementary material

.