Optimizing Colorectal Cancer Screening Strategies and Target Populations in Budget-Constrained Regions through Cost-Effectiveness Analysis: A Case from China's East

doi:10.21203/rs.3.rs-3217334/v1

Download PDF

Research Article

Optimizing Colorectal Cancer Screening Strategies and Target Populations in Budget-Constrained Regions through Cost-Effectiveness Analysis: A Case from China's East

https://doi.org/10.21203/rs.3.rs-3217334/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Introduction: Colorectal cancer (CRC) is among the top three cancers globally in terms of morbidity and mortality. For middle-income countries planning to implement regional screening programs, conducting a preliminary assessment to identify individuals at a higher risk of CRC before proceeding with a colonoscopy can be more cost-effective.

Methods: We employed a microsimulation model to assess the impact of various preliminary assessment strategies. Simulations were conducted to identify the optimal age for initiating and ending screening.

Results: The primary data were sourced from the Huzhou screening program, which included 418,805 individuals from 2020 to 2022. All screening strategies were found to be effective, with the cost per incremental Quality-Adjusted Life Year (QALY) being less than $1,026, which is below the minimum standard for upper-middle-income countries. The most effective screening strategy was the annual combined two-sample Immunochemical Fecal Occult Blood Test (i-FOBT) and risk evaluation questionnaires. This approach led to a reduction in CRC incidence and related deaths by 2435 and 1174 cases per 100,000 individuals, respectively, and an increase in Life Years Saved (LYS) by 13903 years and QALYs by 35564 years. The recommended ages to begin and end screening were 48 and 72 years, respectively.

Conclusions: All CRC screening strategies demonstrated effectiveness compared to non-screening, with the annual combined two-sample i-FOBT and risk evaluation questionnaires emerging as the optimal approach. For additional regions, the best screening strategy can be selected based on the health outcomes and costs we have provided.

Colorectal Cancer

Microsimulation model

Health Economics

Early Detection of Cancer

What is already known on this topic:

According to CI5 (Cancer Incidence in Five Continents), colorectal cancer (CRC) incidence rates are higher in most European and American countries than in Asian and African countries.
Implementing region-wide screening programs can slow down the incidence rate in high-income countries, yet the incidence rate continues to rise rapidly in Asian and African countries that have yet to conduct regional s.
Implementing a preliminary assessment before colonoscopy appears to be a more appropriate strategy, particularly in less developed regions where the cost of colonoscopy-based screening is relatively high.

What this study adds:

In this study, various preliminary assessment methods and implementation frequencies for colorectal cancer (CRC) were compared, leading to the identification of the most suitable screening strategy for middle-income countries.
This study found that a combination of a risk evaluation questionnaire and two samples i-FOBT for preliminary assessment, followed by colonoscopy for high-risk patients, yielded the best health benefits as a screening strategy, given that the incremental costs are acceptable.
The optimal starting age and stopping age for the implementation of colorectal cancer screening program are 48 and 72 years, respectively.

How this study might affect research, practice or policy:

The study's results have important implications for middle-income countries that are planning to implement CRC screening programs.
In high-income countries, the common screening approach for colorectal cancer involves direct colonoscopy-based examination and fecal DNA testing. However, our research indicates that the recommended preliminary assessment strategy, which includes a risk evaluation questionnaire and i-FOBT (immunochemical fecal occult blood test), can achieve very high sensitivity. Moreover, this method is less invasive and more cost-effective.
Furthermore, this study provides rich insights into different strategies, offering valuable perspectives for the implementation of screening programs in relevant countries and regions.

Colorectal cancer (CRC) is one of the most common cancers and one of the most frequent causes of cancer deaths worldwide¹. Because CRC is not associated with any specific symptoms, most patients are diagnosed with advanced-stage disease ². The 5-year survival rate and quality of life of patients with CRC are closely related to tumor stage at initial diagnosis, with 5-year survival rates ranging from 90% in patients with polyps to 14% in patients with late-stage CRC³. Early detection and access to treatment are therefore critical. Moreover, due to the slow progression of this disease, the probability of detecting the disease increases significantly with regular screening. Consequently, screening for CRC has been regarded as a cost-effective tool to reduce the disease burden^3-5.

Research by Arnold and colleagues has shown that with sustained, long-term CRC screening, high screening coverage, and advanced diagnostic and treatment methods, both incidence and mortality rates can be reduced ⁶. For example, in countries such as the United States, Japan and France. If there are well-established healthcare resources providing screening services and standardized diagnostic and treatment techniques, incidence rates may continue to rise but mortality rates remain stable or even decline in countries such as Canada, the United Kingdom, Italy, Denmark and Singapore. However, in other countries such as Russia, China and Brazil as well as most Asian and African countries both incidence and mortality rates are on the rise. In the near future, an increasing number of upper-middle-income countries (with a per capita GNI ranging from $1,026 to $12,535), like China, will begin to implement nationwide screening programs. However, currently, there are relatively few studies conducted on this topic in these countries, leading to a lack of reliable references for guidance. Data from China’s screening programs can provide valuable insights for similar countries in selecting appropriate screening strategies and target populations.

stool DNA (sDNA) tests, guaiac fecal occult blood tests (gFOBT), and Immunochemical Fecal Occult Blood Test (iFOBT)7-9. The usual recommendation is to conduct gFOBT or iFOBT^7-9. The usual recommendation is to conduct gFOBT and iFOBT, which detect potential bleeding in the stool, on an annual or biennial basis. Due to its instability and potential for false positive results depending on diet, gFOBT is generally not recommended for routine use¹⁰. On the other hand, iFOBT, which detects antibodies to globin, can provide more reliable results. Despite sDNA tests being more sensitive than FOBT and based on gene sequencing, they are rarely utilized in developing countries with limited budgets due to their significantly higher costs. These aforementioned methods are typically employed as initial screening for the target population before undergoing colonoscopy. However, in some developed countries, colonoscopy or sigmoidoscopy is frequently performed directly without preliminary assessment¹¹.

Utilizing data from the Huzhou City CRC screening project, the Health Insurance Bureau, and other literature sources, we simulated the health benefits and economic costs of different screening strategies using the Colon Modeling with Open Source Tool (CMOST)¹². Although current screening technologies are relatively mature, it is challenging to determine the most cost-effective and suitable screening strategies for middle-income countries and regions. The disease parameters in this model were obtained from field screening, which can reflect the disease situation of the population and provide a basis for determining the disease burden of CRC. These findings may enable improvements in screening middle-aged and elderly Chinese subjects for CRC. Moreover, screening methods in China can be used to design large-scale screening programs in other countries and regions.

Study design

Our study is a cost-effectiveness analysis from a societal perspective, using a microsimulation model to obtain population health outcomes and costs.

Data sources

The data for this study were sourced from various datasets and resources, including: (1) Screening data from Huzhou from 2020 to 2022 were used to obtain relevant parameters for different screening strategies. The data also provided detection rates for various lesions at different ages, which helped to calibrate the model. Individual participation rates in the screening and colonoscopy procedures were also obtained from this dataset (the screening process is provided in Figure S1). (2) Data from the seventh census of China, Huzhou, were used to construct life tables (Table S1). (3) Huzhou Statistics Yearbook: This official report provided fundamental information about Huzhou City, including demographic and socio-economic data. (4) Zhejiang Provincial Medical Security Bureau data: This dataset was primarily used to acquire treatment costs and expenses related to diseases. (5) Other sources: Certain aspects of the disease's natural history relied on disease progression data predefined in the model, moreover, some parameters were obtained from literature. For a detailed list of data sources, please refer to Supplementary Table S2.

Setting and location

The setting for this study is Huzhou City, Zhejiang Province, China. The total population of Huzhou is 2.68 million, with the Han ethnic group accounting for 96.57% of the population. The overall sex ratio (calculated as the number of males per 100 females) is 109.04 and 18.70% of the population is aged 60 or above. The per capita GDP of Huzhou is approximately $15,630. The number of healthcare workers in Huzhou is approximately 11.74 per thousand people and the number of hospital beds is approximately 6.37 per thousand people. According to data from 14 cancer registries in Zhejiang Province, the incidence rate of colorectal cancer is approximately 41.75 per 100,000 people and the mortality rate due to colorectal cancer is approximately 17.02 per 100,000 people.

Screening strategies and Sensitivity of screening methods

The initial screening process, prior to recommending colonoscopy, involves an evaluation of four distinct screening methods, each with two screening frequencies: annual and biennial. This results in a total of eight screening strategies. The four methods are as follows: (1) Method 1, Single-sample iFOBT: A positive result in the iFOBT indicates a positive screening. (2) Method 2, Two-sample iFOBT (with a one-week interval between samples): Any positive iFOBT result from either of the two samples indicates a positive screening. (3) Method 3, Single-sample iFOBT and risk evaluation questionnaire: A positive screening is determined if the iFOBT result is positive or if the risk evaluation questionnaire categorizes the individual as high risk. (4) Method 4, Two-sample iFOBT and risk evaluation questionnaire: A positive screening is indicated if either of the iFOBT samples is positive or if the risk evaluation questionnaire categorizes the individual as high risk.

This study utilized the iFOBT method, employing the latex agglutination turbidimetric technique. The latex agglutination turbidimetric technique allows for the quantitative measurement of low concentrations of hemoglobin in feces. A positive result was defined as a hemoglobin content of 20μg per gram of formed feces. To address potential false-positive results caused by conditions such as hemorrhoids bleeding, menstrual bleeding in females, gingival bleeding, or gastrointestinal ulcers, the iFOBT was deferred. For the Two-sample iFOBT method, the two samples were collected with a 7-day interval between them.

According to the Chinese colorectal cancer screening guidelines¹³, the following items were selected as risk assessment items: history of colorectal polyposis, familial adenomatous polyposis, age, sex, family history of colorectal cancer, smoking, and body mass index (Table 1). Each item was meta-analyzed and was considered to be effective. The risk assessment questionnaire will be used as an auxiliary assessment tool with iFOBT to form a primary screening model.

Table 1 Risk evaluation questionnaire

Risks	Group	Score	Estimate the results
History of colorectal polyposis			High risk
Familial adenomatous polyposis			High risk
Age (years)	50-54	0	Total score ≥ 5 points, High-risk 1-4 points, Medium-risk 0 points, Low-risk
	55-64	1
	65-74	2
Sex	Female	0
Sex	male	1
Family history of colorectal cancer	No	0
Family history of colorectal cancer	Yes	1
Smoking	Non-smoking	0
Smoking	Smoking	1
Body mass index (BMI, kg/m²)	＜23	0
Body mass index (BMI, kg/m²)	≥ 23	1

Sensitivity and specificity of screening methods

To calculate the sensitivity and specificity of different screening methods, we first estimated the number of patients. We assumed that the probability of an individual undergoing colonoscopy is random if the preliminary screening results are consistent (the results of the first iFOBT, the second iFOBT, and the risk assessment questionnaire are all in agreement). In this case, the prevalence of disease in individuals who undergo colonoscopy and those who refuse colonoscopy is the same. The calculation process is shown in Table S3. We separately calculated the detection patients of the four screening methods. The sensitivity and specificity were then calculated.

Health outcomes

The health outcomes of this study include the number of colorectal cancer cases, the number of deaths from colorectal cancer, the years of life lost due to colorectal cancer, and the quality-adjusted years of life lost due to colorectal cancer. The first three outcomes can be directly obtained from the model, while QALYs take into account both the duration spent in different health states and the quality of life associated with each state. Quality of life is measured on a scale from 0 to 1, where 0 represents death and 1 represents perfect health, as presented in Table S2^14,15.

Cost

The cost estimation for the screening in this study comes from the actual expenditure of the colorectal cancer screening program in Huzhou City. The costs of colonoscopy, precancerous adenoma treatment, treatment for colonoscopy bleeding, treatment for colonoscopy perforation, and treatment for early and advanced CRC are all derived from data provided by the Zhejiang Provincial Medical Insurance Bureau. All costs considered are direct medical costs, with indirect medical costs and direct non-medical costs not taken into account in this study. To enhance the promotional viability of this study, the monetary unit employed is the US dollar. All cost data referenced in this paper have been converted from Chinese Yuan to US dollars, utilizing the exchange rate of 1 US dollar to 6.88 Chinese Yuan as of March 17th, 2023.

Cost- effectiveness analysis

In the cost-effectiveness analysis phase, the health outcome measures considered in this study include LYs and QALYs. We computed the Incremental Cost required for each additional LY or QALY gained by different screening strategies, compared to the no screening option. Incremental Cost- effectiveness Ratio (ICER) is defined as the increase in cost for each LYs or QALYs obtained, including costs of CRC screening and treatment, and has been used to compare different choices.

Model

CMOST is an open-source and reliable model used for simulating the progression of colorectal cancer and estimating the impact of different colorectal screening strategies on disease burden and health outcomes¹². It employs discrete event modeling, which offers greater flexibility and generates simulation results that closely resemble real-world scenarios compared to traditional decision tree Markov models. However, its use requires more detailed parameter support. ¹⁶. In this study, CMOST was employed to simulate the disease burden and health outcomes for 100,000 residents aged 45 to 100 years under four screening methods and two screening frequencies, resulting in a total of eight scenarios. The screening age range was set between 45 and 75 years. Based on survey findings, the screening acceptance rate was defined as 0.57, and the probability of individuals proceeding to subsequent colonoscopy after a positive preliminary screening was set at 0.46. The sensitivity and specificity calculations for different screening strategies can be found in Table 2, The simulation of the diseases natural history is individual-based and calibrated according to the age-specific detection rates of precancerous lesions and cancer obtained from Huzhou Screening program results.

The screening process is set up so that patients who are initially screened positive will continue to be recommended to undergo colonoscopy. Individuals will no longer be screened in the future five years after undergoing colonoscopy. Patients diagnosed with precancerous lesions will receive treatment. According to the recommendations of the WHO and the Chinese Guidelines for the Treatment of Colorectal Cancer, precancerous lesions that are early adenomas will be monitored once every five years after treatment, and precancerous lesions that are advanced adenomas will be detected once in the third year after treatment, and then monitored once every five years. Patients who are diagnosed with colorectal cancer will be monitored once each in the first year and the fourth year, and then monitored once every five years. For a comprehensive understanding of the models, please consult the CMOST manual.

The probability that a person with a given set of preliminary assessment results was in a given category was determined from the actual results of the screening project. The proportion of individuals with CRC during each period was not altered by the decision to undergo or not undergo colonoscopy. All costs and effects were calculated at a discount rate of 5% per year, as recommended by the China Guidelines for Pharmacoeconomic Evaluations¹⁷. The initial screening process, prior to recommending colonoscopy, involves an evaluation of four distinct screening methods, each with two screening frequencies: annual and biennial. This results in a total of eight screening strategies.

Sensitivity analysis

In the sensitivity analysis, we focus on the sensitivity and specificity of the CRC screening, the screening acceptance rate, the colonoscopy acceptance rate, as well as the impact of screening and colonoscopy costs on Incremental Lys, Incremental QALYs, and Incremental costs. The range settings for these parameters used in the sensitivity analysis are presented in the results section.

The microsimulation process was implemented using MATLAB (MATLAB Release 2023a), and data visualization was performed using R 4.3.1 (R Foundation for Statistical Computing, Vienna, Austria).

Basic information

In the first phase of the screening program in Huzhou City, the primary objective was to conduct a one-time screening for 967,234 individuals within a five-year period. Over the course of the Huzhou City screening project from 2020 to 2022, a total of 732,531 individuals were included in the target population, with no overlap between the target populations in each year. Eventually, 418,805 individuals actively participated in the screening, resulting in a screening acceptance rate of 57.2%. Huzhou City implemented the Two-sample iFOBT and risk evaluation questionnaire screening plan. During the initial screening stage, 84,545 individuals tested positive. Following recommendations and mobilization efforts, a total of 38,870 individuals, accounting for 46.0%, underwent colonoscopy.

Comparison of Different Screening Strategies

Through simulation of 100,000 individuals from ages 45 to 100, with the screening age range set between 45 and 75, we compared eight different screening strategies. Overall, all screening strategies were found to be effective compared to the no screening option. Regarding the screening methods, the combination screening approach (i.e., considering any positive result from initial screening as a positive case) led to increased sensitivity with the addition of screening methods, but a decrease in specificity (Table 2). When employing the same screening method, annual implementation yielded higher LYs and QALYs compared to biennial implementation. Among the strategies, the combined use of Single-sample iFOBT and risk evaluation questionnaire with biennial implementation was found to be the most cost-effective for gaining an incremental LY, with a cost of $1,609. On the other hand, Single-sample iFOBT with annual implementation was identified as the most cost-effective for gaining an incremental QALY, with a cost of $635. In the absence of cost considerations, conducting annual Two-sample iFOBT with risk evaluation questionnaire yielded the greatest health outcome benefits, leading to a reduction of 2,435 cases of CRC incidence and 1,174 deaths among 100,000 individuals. The incremental LYs were 13,903 years, and the additional QALYs amounted to 35,564 years. When the government covers the cost of initial screening and individuals bear the subsequent colonoscopy expenses, the government would need to allocate approximately $3,654,745 for the initial screening, while each individual would incur an average cost of approximately $35.6 for the follow-up colonoscopy. For additional findings, please refer to Table 3.

Since the cost for gaining an incremental QALYs in all strategies is below $1,026, which is lower than the World Bank's threshold for Upper-middle-income countries, we designate the best screening strategy as the Two-sample iFOBT with risk evaluation questionnaire, implemented annually. We applied this strategy in subsequent simulations for different starting and stopping ages, as well as for screening acceptance rate analysis and sensitivity analysis. As the differences among screening strategies mainly lie in sensitivity, specificity, and screening costs, and screening does not affect the natural history of the disease, the results of simulations for different starting and stopping ages, screening acceptance rates, and sensitivity analyses can be considered largely consistent across different strategies.

Table 2 Specificity and Sensitivity of Different Screening Strategies

Screening Strategies	Population of Screening Positivity	Specificity	Sensitivity of Precancerous Lesions		Sensitivity of CRC
Screening Strategies	Population of Screening Positivity	Specificity	Early Adenoma	Advanced Adenoma	Early	Advanced
Method 1	1st Sample i-FOBT Pos.	0.905	0.091	0.254	0.555	0.790
Method 2	≥1 Sample(s) i-FOBT Pos.	0.872	0.112	0.331	0.667	0.905
Method 3	1st Sample i-FOBT Pos. or Qnr. HR	0.827	0.130	0.430	0.917	0.894
Method 4	≥1 Sample(s) i-FOBT Pos. or Qnr. HR	0.797	0.147	0.497	0.999	0.999

Abbreviations: CRC : Colorectal Cancer; i-FOBT: Immunochemical Fecal Occult Blood Test; Pos.: Positive; Qnr.: Questionnaire; HR : High Risk

Table 3 Evaluations of Different Screening Strategies

Item	No Screening	Method 1^s		Method 2		Method 3		Method 4
Item	No Screening	annual	biennial	annual	biennial	annual	biennial	annual	biennial
Overall	100000
Screening Acceptance Rate	0.57
Colonoscopy Acceptance Rate	0.46
Screening Counts	0	1007273	736533	901908	687925	792666	627187	730949	590418
Colonoscopies
Screening	0	113336	81822	139430	103940	166628	129158	181786	143507
Symptomatic	4997	2237	2817	2010	2444	1747	2103	1656	1859
Follow-up	13472	56718	49022	64522	54640	69407	61157	72730	65075
Overall	18469	172291	133661	205962	161024	237782	192418	256172	210441
Increased Counts		158819	120189	192490	147552	219313	178946	242700	196969
Adverse reaction of Colonoscopy
Slight bleeding (No treatment needed)	305	2841	2204	3396	2655	3921	3173	4224	3470
Severe bleeding	1	9	7	11	8	12	10	13	11
Perforation	0	4	3	5	4	6	5	6	5
Population of CRC
Screening	0	716	640	755	723	713	652	716	718
Symptomatic	4986	2228	2809	2007	2434	1740	2096	1655	1857
Follow-up	46	201	156	233	199	239	222	226	240
Overall	5032	3145	3605	2995	3356	2692	2970	2597	2815
Increased Counts		-1887	-1427	-2037	-1676	-2340	-2062	-2435	-2217
Proportion of CRC stage
Stage I	15.8	23.3	22.3	24.7	23.8	27.5	25.6	28.1	25.8
Stage II	36.6	37.2	39.1	36.9	37.8	34.9	35.0	36.0	37.5
Stage III	27.2	23.1	23.4	24.3	23.9	22.7	23.9	21.2	22.0
Stage IV	20.4	16.4	15.3	14.0	14.5	14.9	15.4	14.7	14.7
Health outcomes
CRC Deaths	2037	1126	1280	1022	1164	878	1003	863	937
Changed Deaths		-911	-757	-1015	-873	-1159	-1034	-1174	-1100
LYs lost to colon cancer	26152	16054	17081	14554	16160	13119	13639	12249	13387
Incremental LYs		10098	9071	11598	9992	13033	12513	13903	12765
Incremental LYs / person		0.101	0.091	0.116	0.100	0.130	0.125	0.139	0.128
QALYs lost to colon cancer	57969	31521	35702	27881	32394	23964	27055	22405	25598
Incremental QALYs		26448	22267	30088	25575	34005	30914	35564	32371
Incremental QALYs / person		0.264	0.223	0.301	0.256	0.340	0.309	0.356	0.324
Cost
Screening	0	2,014,546	1,473,066	3,607,632	2,751,700	2,377,998	1,881,561	3,654,745	2,952,090
Screening colonoscopy and Adverse reaction	0	22,213,856	16,037,112	27,328,280	20,372,240	32,659,088	25,314,968	35,630,056	28,127,372
Total Screening Cost	0	24,228,402	17,510,178	30,935,912	23,123,940	35,037,086	27,196,529	39,284,801	31,079,462
Other colonoscopy and Adverse reaction	3,619,983	11,556,379	10,161,362	13,041,716	11,189,581	13,946,184	12,400,306	14,581,466	13,120,541
Precancerous Lesions and CRC	120,628,041	105,282,568	112,018,185	108,391,086	108,422,422	104,380,507	104,788,461	103,093,107	103,604,843
Total	124,248,024	141,067,350	139,689,725	152,368,714	142,735,944	153,363,777	144,385,296	156,959,374	147,804,846
Cost / person	1242	1411	1397	1524	1427	1534	1444	1570	1478
Incremental costs per person		168	154	281	185	292	201	327	236
Strategies Evaluation
Incremental costs/ LYs		1666	1702	2425	1850	2243	1609	2353	1845
Incremental costs/ QALYs		635	692	934	723	858	650	919	729

Abbreviations: CRC : Colorectal Cancer; LYs: Life-Years; QALYs: Quality-Adjusted Life-Years; Method 1, Single-sample iFOBT; Method 2, Two-sample iFOBT; Method 3, Single-sample iFOBT and risk evaluation questionnaire; Method 4, Two-sample iFOBT and risk evaluation questionnaire.

Starting and stopping ages

By setting the stopping age at 74 years and simulating different starting ages, the scope of screening becomes narrower as the starting age increases. Consequently, individual incremental LYs, incremental QALYs, and incremental Cost all exhibit a decreasing trend. Based on observations from the graphs, we believe that the optimal starting age is 48 years. Before the age of 48, LYs and QALYs remain relatively stable, and thereafter, the decline accelerates, and the incremental Cost also indicates 48 years as the optimal screening starting age. Similarly, by setting the starting age at 45 years and simulating different stopping ages, the scope of screening expands as the stopping age increases. As a result, individual incremental LYs, incremental QALYs, and incremental Cost generally show an increasing trend. From the observed graphs, we determine that the optimal stopping age is 72 years. Beyond the age of 72, the average individual incremental LYs and incremental QALYs no longer increase, and the incremental Cost also indicates 72 years as the optimal screening stopping age, shown in Figure 1.

The figure illustrates the changes in health outcomes and costs resulting from different starting and stopping ages in a simulated population. Incremental LYs (Life-Years) represent the additional years of life gained through screening, while Incremental QALYs (Quality-Adjusted Life-Years) measure the additional years of life adjusted for the quality of life experienced. Incremental Cost is presented in USD (United States Dollars) and reflects the additional cost incurred for implementing the screening strategies.

Sensitivity analysis

In the sensitivity analysis, we considered the sensitivity of early-stage CRC (0.5 to 1.0), the sensitivity of advanced-stage CRC (0.5 to 1.0), the specificity of screening (0.5 to 1.0), the screening acceptance rate (0.2 to 1.0), the subsequent colonoscopy acceptance rate (0.2 to 1.0), and their impacts on LYs lost to CRC and QALYs lost to CRC. Additionally, we explored the effects of screening costs and colonoscopy costs on the total cost of CRC. The results are presented in Figure 2.

The figure illustrates the changes in Life-Years (LYs) lost to colorectal cancer (CRC), Quality-Adjusted Life-Years (QALYs) lost to CRC, and the total cost of CRC in the simulated population as different parameters vary during sensitivity analysis. The costs are presented in United States Dollars (USD).

This study utilized a microsimulation model to simulate the health outcomes and costs for 100,000 individuals under different screening strategies. All screening approaches were found to be effective, and the cost required to gain an incremental QALY was less than $1,026 (the minimum standard for upper-middle-income countries according to the World Bank). Therefore, the optimal strategy would be the one that maximizes health outcomes. In this case, the combination of two-sample iFOBT and risk assessment questionnaire was found to be the most favorable approach, with annual screening yielding the best health outcomes, including reduced cases of CRC incidence, CRC-related deaths, LYS, and QALYs. Additionally, based on our findings, we recommend a screening starting age of 48 and a screening stopping age of 72.

There are primarily two methods for colorectal cancer screening: stool-based tests and direct visualization tests. Reports from the United States and the European Union indicate that a significant proportion of the population undergoes direct visualization screening, which includes CT colonography, sigmoidoscopy, and colonoscopy, with screening frequencies set at every five or ten years. According to national cancer registry data, this screening method has shown certain effectiveness^10,11. However, considering cost and acceptance rates, it is advisable for high and middle-income countries to implement stool-based screening when initiating screening programs. Stool-based screening primarily includes gFOBT, iFOBT, and stool DNA-FIT¹⁸. However, the stability of gFOBT has been challenged, and the cost of stool DNA-FIT is high. Among these, the immunochemical-based iFOBT is the most mature and widely used, and its sensitivity has been proven to be relatively stable through numerous studies. Therefore, our research recommends its use as the primary method for initial screening.

We developed our risk assessment questionnaire in line with the Chinese guidelines for colorectal cancer screening and with further reference to the recommendations of the Asia Pacific Working Group^13,19. Our results indicate that when the risk assessment questionnaire is used in conjunction with iFOBT, there is a significant increase in the proportion of early colorectal cancer cases in the overall patients, suggesting the effectiveness of the questionnaire. Guidelines vary across countries, reflecting local conditions. For instance, most guidelines, such as those from the American Society of Gastroenterology and the U.S. Multi-Society Task Force, regard family history as an important risk factor and classify subjects without a family history of CRC as being at general risk^20,21. In contrast, the European CRC screening guidelines Working Group classifies individuals with a family history of CRC, but not with a relevant genetic syndrome, as being at general risk for CRC²². Therefore, we propose that when the combined screening strategy of iFOBT and questionnaire is used in more countries, our questionnaire should be modified according to the guidelines of the respective country.

Field surveys suggest that approximately 57% of patients are willing to undergo screening, and about 46% of patients are willing to proceed with a colonoscopy following a positive screening result. This significantly contrasts with the over 90% acceptance rate reported in countries like Italy, which may be attributed to differences in economic levels, healthcare resources, and the duration of screening implementation¹¹. In another study on the initial implementation of colorectal cancer screening in Spain, we found acceptance rates similar to ours²³. We believe that the acceptance rate for screening will increase over time. In our sensitivity analysis, we simulated scenarios with varying screening and colonoscopy acceptance rates, which allowed us to estimate the health outcomes and costs should the acceptance rates increase.

In this study, we recommend the optimal starting and stopping ages for screening as 48 to 72 years. However, this recommendation is made at the national level for newly initiated screenings. Although the prevalence of early-onset CRC is low, its progression is rapid²⁴. As the program is implemented, screening strategies for younger populations can be further optimized. Similarly, for the elderly population, the incidence of adverse events from colonoscopy increases, and the benefits of screening are controversial, which may require more personalized approaches²⁵.

Despite the comprehensive findings, there are several limitations to this study. Firstly, the parameters for the state transition model were not available from our survey, so we used the default parameters in CMOST. These parameters might differ from the actual situation ¹². However, these parameters were obtained from numerous studies and have been tested for reliability²⁶. Moreover, we calibrated the state transition model based on the detection rates for each age and gender in our study. Secondly, it is difficult to ascertain the true disease status of individuals who are at low risk and have two negative iFOBT results, as they are not recommended for colonoscopy. Among the 81 eligible individuals who underwent colonoscopy, no cases of CRC were detected, leading to an estimated CRC incidence of 0 for this subgroup. This could lead to an underestimation of the total number of patients and an overestimation of the sensitivity of the screening strategy. However, our estimated sensitivities for single-sample iFOBT screening for early and advanced CRC were 0.56 and 0.76, respectively. In comparison, the reference range given by the American working group, without distinguishing between early and late stages, is 0.64 to 0.83, and the Chinese working group's reference range is 0.76 to 0.88. This suggests that our sensitivity estimates are not overestimated.

In conclusion, we find all the examined screening strategies to be effective and their costs manageable for all upper-middle-income countries. During implementation, different countries and regions can modify the questionnaire according to their own circumstances and choose the appropriate screening strategy. The recommended starting and stopping ages for screening are 48 and 72 years respectively. As the screening program progresses, this range can be further expanded.

a. Competing interests

The authors declare that they have no known competing financial interests or personal relationships that could influence the work reported in this paper.

b. Ethics approval and consent to participate

In our study, ethical considerations were in accordance with the Helsinki Declaration. The study protocol was reviewed and approved by the Institutional Review Board (IRB) with the approval number IRB # 2022-12-1019, and the study was also registered internationally with the IRB registration number IRB00002408 & FWA00002399.

c. Consent for publication

Not applicable.

d. Availability of data and materials

The parameters used in the model are provided in Tables. Any additional raw data used in this study are available from the corresponding author upon request.

e. Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

f. Authors' contributions

Hao Li contributed to the formal analysis, investigation, methodology, software, visualization, and writing the initial draft. Yun Fu contributed to data curation, investigation, validation, and writing the initial draft. Ao Xu contributed to the investigation, methodology, software, and writing the initial draft. Wei-bing Wang and Peng Zhang contributed to the conceptualization, project administration, resources, supervision, validation, review, and editing of the manuscript. Wei-bing Wang also contributed to funding acquisition.

g. Acknowledgements

We gratefully acknowledge our funders and the data provided by the Huzhou Center for Disease Control and Prevention.

Sung H, Ferlay J, Siegel RL, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin May. 2021;71(3):209–49. 10.3322/caac.21660.
Dekker E, Tanis PJ, Vleugels JLA, Kasi PM, Wallace MB. Colorectal cancer. Lancet. Oct 19 2019;394(10207):1467–1480. 10.1016/s0140-6736(19)32319-0.
Siegel RL, Miller KD, Goding Sauer A, et al. Colorectal cancer statistics, 2020. CA Cancer J Clin May. 2020;70(3):145–64. 10.3322/caac.21601.
Levin TR, Corley DA, Jensen CD, et al. Effects of Organized Colorectal Cancer Screening on Cancer Incidence and Mortality in a Large Community-Based Population. Gastroenterol Nov. 2018;155(5):1383–1391e5. 10.1053/j.gastro.2018.07.017.
Pilonis ND, Bugajski M, Wieszczy P, et al. Participation in Competing Strategies for Colorectal Cancer Screening: A Randomized Health Services Study (PICCOLINO Study). Gastroenterol Mar. 2021;160(4):1097–105. 10.1053/j.gastro.2020.11.049.
Arnold M, Sierra MS, Laversanne M, Soerjomataram I, Jemal A, Bray F. Global patterns and trends in colorectal cancer incidence and mortality. Gut Apr. 2017;66(4):683–91. 10.1136/gutjnl-2015-310912.
Lin JS, Piper MA, Perdue LA, et al. Screening for Colorectal Cancer: Updated Evidence Report and Systematic Review for the US Preventive Services Task Force. Jama Jun. 2016;21(23):2576–94. 10.1001/jama.2016.3332.
Bénard F, Barkun AN, Martel M, von Renteln D. Systematic review of colorectal cancer screening guidelines for average-risk adults: Summarizing the current global recommendations. World J Gastroenterol Jan. 2018;7(1):124–38. 10.3748/wjg.v24.i1.124.
Schreuders EH, Ruco A, Rabeneck L, et al. Colorectal cancer screening: a global overview of existing programmes. Gut Oct. 2015;64(10):1637–49. 10.1136/gutjnl-2014-309086.
Haghighat S, Sussman DA, Deshpande A. US Preventive Services Task Force Recommendation Statement on Screening for Colorectal Cancer. JAMA Oct. 2021;5(13):1328. 10.1001/jama.2021.13466.
Senore C, Basu P, Anttila A, et al. Performance of colorectal cancer screening in the European Union Member States: data from the second European screening report. Gut Jul. 2019;68(7):1232–44. 10.1136/gutjnl-2018-317293.
Prakash MK, Lang B, Heinrich H, et al. CMOST: an open-source framework for the microsimulation of colorectal cancer screening strategies. BMC Med Inform Decis Mak Jun. 2017;5(1):80. 10.1186/s12911-017-0458-9.
Wanqing C, Ni L, Ping L, et al. China Guideline for the Screening, Early Detection and Early Treatment of Colorectal Cancer (2020, Beijing). Chin Cancer. 2021;30(01):1–28.
Huang W, Yu H, Liu C, et al. Assessing Health-Related Quality of Life of Chinese Adults in Heilongjiang Using EQ-5D-3L. Int J Environ Res Public Health Feb. 2017;23(3). 10.3390/ijerph14030224.
Huang W, Yang J, Liu Y, et al. Assessing health-related quality of life of patients with colorectal cancer using EQ-5D-5L: a cross-sectional study in Heilongjiang of China. BMJ Open Dec. 2018;6(12):e022711. 10.1136/bmjopen-2018-022711.
Deibel A, Deng L, Cheng CY, et al. Evaluating key characteristics of ideal colorectal cancer screening modalities: the microsimulation approach. Gastrointest Endosc Aug. 2021;94(2):379–390e7. 10.1016/j.gie.2021.02.013.
Association CP. Chapter CSfPaORIB. China Guidelines for Pharmacoeconomic Evaluations. https://tools.ispor.org/peguidelines/source/China-Guidelines-for-Pharmacoeconomic-Evaluations-2020.pdf.
Lin JS, Perdue LA, Henrikson NB, Bean SI, Blasi PR. Screening for Colorectal Cancer: Updated Evidence Report and Systematic Review for the US Preventive Services Task Force. JAMA May. 2021;18(19):1978–98. 10.1001/jama.2021.4417.
Sung JJ, Ng SC, Chan FK, et al. An updated Asia Pacific Consensus Recommendations on colorectal cancer screening. Gut Jan. 2015;64(1):121–32. 10.1136/gutjnl-2013-306503.
Rex DK, Johnson DA, Anderson JC, Schoenfeld PS, Burke CA, Inadomi JM. American College of Gastroenterology guidelines for colorectal cancer screening 2009 [corrected]. Am J Gastroenterol Mar. 2009;104(3):739–50. 10.1038/ajg.2009.104.
Rex DK, Boland CR, Dominitz JA, et al. Colorectal Cancer Screening: Recommendations for Physicians and Patients from the U.S. Multi-Society Task Force on Colorectal Cancer. Am J Gastroenterol Jul. 2017;112(7):1016–30. 10.1038/ajg.2017.174.
Malila N, Senore C, Armaroli P. European guidelines for quality assurance in colorectal cancer screening and diagnosis. First Edition–Organisation. Endoscopy Sep. 2012;44(Suppl 3):e31–48. 10.1055/s-0032-1309783.
Quintero E, Castells A, Bujanda L, et al. Colonoscopy versus fecal immunochemical testing in colorectal-cancer screening. N Engl J Med Feb. 2012;23(8):697–706. 10.1056/NEJMoa1108895.
Cheng E, Blackburn HN, Ng K, et al. Analysis of Survival Among Adults With Early-Onset Colorectal Cancer in the National Cancer Database. JAMA Netw Open Jun. 2021;1(6):e2112539. 10.1001/jamanetworkopen.2021.12539.
Bhoo-Pathy N, Bujang NN, Ng CW. Continuation of Screening Endoscopy for Colorectal Cancer in Older Adults. JAMA Oncol Jul. 2021;1(7):973–5. 10.1001/jamaoncol.2021.1119.
Vivas-Valencia C, Zhou Y, Sai A, Imperiale TF, Kong N. A two-phase approach to re-calibrating expensive computer simulation for sex-specific colorectal neoplasia development modeling. BMC Med Inform Decis Mak Sep. 2022;18(1):244. 10.1186/s12911-022-01991-7.

No competing interests reported.

Supplements.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

Optimizing Colorectal Cancer Screening Strategies and Target Populations in Budget-Constrained Regions through Cost-Effectiveness Analysis: A Case from China's East

Status:

Version 1

Abstract

Figures

Key questions

Introduction

Methods

Results

Discussion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1