Identification of the implementation influencing factors of the organized breast cancer screening in Ordos city: from caregivers’ perspectives

Abstract

Background

An organized breast cancer screening program has been introduced by the local government in Ordos City since 2010. Women between 35 and 64 years are eligible to be screened freely at local maternal and child healthcare hospitals. The screening coverage rate in Ordos City is relatively higher than the national average level. However, the overall breast cancer detection rate remains low. This study aims to identify contextual influencing factors of this program from local maternal and child healthcare hospitals’ perspectives and provide suggestions for screening service improvement in Ordos City.

Methods

We conducted a qualitative study to evaluate the implementation of the organized breast cancer screening program from maternal and child healthcare hospitals’ perspectives in Ordos City. Four domains (outer setting, inner setting, individuals, and process) and nine constructs were selected for evaluation under the instruction of the Consolidated Framework for Implementation Research (CFIR). CIFR constructs were rated in both valence and strength factors. The interview scripts were coded and themes were generated by using content analysis. Data obtained were interpreted and described by tables and charts.

Results

A total of 28 individuals participated in our study. We found that the implementation of the organized breast cancer screening in Ordos City went well. Policies & laws, partnerships & connections, performance-measurement pressure, implementation leads, teaming, and engaging constructs holistically facilitated the implementation performance. Available resources varied and this construct chiefly displayed a neutral impact. Access to knowledge & information and doing constructs were considered to negatively influence the implementation performance. Political commitment to cervical cancer elimination, multisectoral collaboration, specialized leadership, and structured screening teams were the main factors promoting the screening progress. Inconsistent referrals, underqualified professional competence of primary health workers, and non-interoperable electronic systems were the main barriers to screening performance.

Conclusions

We recommend that Ordos City further facilitates the prompt referral for imaging and biopsy, promotes the capacity building of primary health workers, and strengthens the construction of electronic systems. Future quantitative studies are necessary to comprehend the knowledge and attitudes of caregivers and target population toward breast cancer screening. Interviews with other stakeholders, such as policymakers and target population, are also needed.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12913-025-13289-1.

Background

Breast cancer is the most common cancer and the leading cause of cancer death among women globally [1]. It ranks first and fifth in cancer incidence and mortality among Chinese women, respectively. In 2022, there were 357,200 new cases of breast cancer and 75,000 deaths in China [2]. Regular mammography screening is effective in reducing breast cancer mortality [3, 4] and has been adopted by many high-income countries [5, 6]. However, mammography screening is not widely available in low- and middle-income countries due to high expense or limited health resources [7, 8]. Some studies have concluded that the effectiveness of ultrasound is comparable to mammography, especially for women with dense breasts [9–11]. More Asian women are higher in breast density [12, 13] and are less sensitive to mammography [14–16]. Moreover, the breast cancer onset in China is considerably younger than that in Western countries [17].

Considering the feasibility of mammography application, such as cost, radiation, and breast density, China is currently implementing an ultrasound-based modality to screen breast cancer [18]. In 2009, China launched an organized cervical and breast cancer (referred to as “two-cancer”) screening program for women aged 35–59 years in rural areas [19]. Its age range was expanded to 35–64 in 2012. National Health Commission (NHC, former Ministry of Health) formulated the screening modality of breast cancer. The primary examination is breast ultrasound (BUS) combined with clinical breast examination (CBE). Any detected abnormality would be referred to a mammography examination (MAM).

Ordos, a city in the Inner Mongolia Autonomous Region, is a multiethnic region dominated by Mongols. The local government has provided women Living in this region with free two-cancer screening services since 2016 [20]. In 2021, they started a new round of screening. Maternal and child healthcare (MCH) hospitals in each district/banner serve as the primary screening locations. The screening modality follows the NHC’s recommendation.

Health outcomes from 2016 to 2020, such as screening coverage rate and breast cancer detection rate, have been calculated and analyzed [21]. The 5-year screening coverage rate (52.24%) was higher than the national level (30.90%) [22], but the overall breast cancer detection rate (98/173,808) was lower than the national level in China (56.38/10⁵ vs. 85.30/10⁵) [9, 21]. This could related to inadequate quality of screening services and follow-up management. Qualitative methods are helpful in deeply understanding the reasons. Therefore, we conducted a qualitative study to assess contextual influencing factors to the implementation of the organized breast cancer screening (BCS) from MCH hospitals’ perspectives. We aim to document suggestions to improve service quality in Ordos City.

Methods

We reported our study following the Consolidated Criteria for Reporting Qualitative Studies (COREQ) checklist (Table S1).

Selection and description of CFIR constructs

The Consolidated Framework for Implementation Research (CFIR) is often used to guide investigators in identifying factors influencing implementation progress and effectiveness. In 2009, Damschroder et al. summarized 19 theories, frameworks, and models applied in health services research, and finally integrated them into a meta-theoretical framework, CFIR [23]. In 2022, Damschroder et al. updated the CFIR to meet the needs of implementation research [24]. Due to its comprehensiveness, flexibility, and accessibility, the CFIR has been widely used in implementation research in areas such as tobacco use [25, 26] weight management [27, 28] and cervical cancer prevention [29, 30] around the world. Its main function is consistent with our study purpose and we selected CFIR to instruct our study design.

CFIR is composed of five Major domains and 48 constructs [24]. We selected four domains (outer setting, inner setting, individuals, and process) and ten constructs (local conditions, policies & laws, partnerships & connections, performance-measurement pressure, available resources, access to knowledge & information, implementation leads, teaming, engaging, and doing) based on our needs. The description of each construct is shown in Table 1. Since local MCH hospitals were basic research units, the inner setting and individuals domains indicated the environment and staff within local MCH hospitals.

Table 1.

The description of CFIR constructs selected

Domains	Constructs	Descriptions
Outer setting	Policies & laws	Are there any supporting policies?
	Partnerships & connections	What are the partnerships of MCH hospitals with external sectors?
	Performance-measurement pressure	How does the completion status of annual screening goals affect the performance of the organized BCS?
Inner setting	Available resources	How many personnel and equipment are available in local MCH hospitals?
	Access to knowledge & information	How much are BCS staff accessible to BCS-related knowledge and information?
Individuals	Implementation leads	How much do the leaders in local MCH hospitals commit to BCS program?
Process	Teaming	How are the screening teams in local MCH hospitals organized?
	Engaging	What has been done to facilitate the participation of target population?
	Doing	How are the organized BCS programs executed? (execution) How are suspected cases followed up? (follow-up) How is the data managed? (data management) Is there any quality control? (supervision)

CFIR the Consolidated Framework for Implementation Research, MCH maternal and child healthcare, BCS breast cancer screening

The outer setting domain articulated the coordination of MCH hospitals with other health or non-health sectors. The inner setting domain elicited available resources in local MCH hospitals, including staff members, equipment, and training and education opportunities. The individuals domain focused on the commitment of leaders in local MCH hospitals to the organized BCS program. The process domain delineated all the aspects of the screening process, including structure of screening teams, engagement of target population, execution of screening work plans, follow-up of suspected cases, and data management.

Study participants and sampling

The basic research units were primary screening sites, local MCH hospitals. Ordos City has nine districts/banners and each district/banner owns one MCH hospital. These nine MCH hospitals undertake the organized BCS program provided by the local government. We recruited people involved in the BCS implementation in local MCH hospitals, including heads of local MCH hospitals, two-cancer screening program managers, directors of the Department of Gynecology, Ultrasound, and Radiology, and other screening-related health workers or administrative staff. Considering that the condition in each district/banner varied, such as population density and land area, each MCH hospital recruited at least one study participant.

Sampling methods included purposive sampling and snowball sampling. All participants were contacted in advance, and were voluntarily participated in this study. The sample size was determined according to the information saturation principle. We stopped the recruitment when no more new information was added.

As for personnel statistics, respondents in this part included medical staff who can conduct CBE/BUS/MAM examination. Anyone who were not responsible for examination was excluded, such as nurses and administrative staff.

Data collection

We collected data from two sources: (1) individual interviews and focus groups, and (2) records and document review.

In-depth interviews

We conducted semi-structured in-depth interviews that were open-ended. Interviews followed the predesigned outline (Table S2). Interview places were interviewees’ offices or private spaces. We required quiet environment to guarantee free expression and avoid interruption. All participants joined our interviews voluntarily. Interviews were carried out in person by two of our research team members (ZC and SW). They were experienced in in-depth interviews, and had conducted mock interviews. One interviewer moderated the process and another interviewer took notes. Meanwhile, we recorded the interview process under the permission of our study participants. Audio documents were transcribed into text documents on the same day of interview by two interviewers. Text documents were then cross-checked by the same interviewers.

Focus groups

In order to improve the efficiency and reduce the inconvenience caused by interviews, focus group discussions were organized when multiple target participants shared common free time. We organized two focus groups (a 6-person group in Junger Banner and a 4-person group in Hangjin Banner). The same interview outline guided the discussions. Discussions were open-ended and participants were free to share their views. We hoped to obtain more information about the experiences and challenges of BCS delivery from their interactions. Two of our research team members (ZC and SW) performed as moderators and note-takers. Discussions were recorded under the permission of our study participants. Audio documents and notes were then transcribed into text documents.

Records and document review

We obtained governmental documents, BCS work plans, and annual screening goals from the Office Department. The actual number of screened women was extracted from the annual screening reports. Information of personnel and equipment (Table S3) was collected directly from the Department of Human Resources, Finance and Equipment. We also photographed equipment for verification with purchase records.

Data analysis

CIFR constructs were rated in both valence and strength factors [31]. Valence evaluated the positive (+), neutral (0) or negative (-) influence of constructs on the implementation performance, and strength evaluated the degree of their influence (1 indicating weak impact and 2 indicating strong impact). We rated nine constructs for each participants and calculated the average score to estimate the overall rating of each construct. Available resources were described using frequencies and rates. The annual screening completion rates in 2021, 2022, and 2023 were calculated based on actual screening number and screening goals.

We use content analysis to process the interview materials. Two of our research team members (ZC and HD) coded the Chinese text documents by NVivo software (version 14). Codes with similar meanings were clustered to generate themes and sub-themes. Themes and sub-themes were translated into English and matched with CFIR constructs. They were then categorized into facilitators or barriers by two coders. Discrepancies were discussed between two coders till reach agreements. Quotations were used to support the interpretation of influencing factors in the results. Partnerships, screening team structure, target population engaging strategies, and screening workflow were displayed by flowcharts or maps.

Results

We conducted this qualitative study from March 2nd to March 31st, 2023. A total of 28 participants were recruited until data saturation. Their information were presented by replacing the name with English letters and numbers (e.g. P1). Personally identifiable information was removed to protect privacy. The participant list is provided in Supplemental file (Table S4). Approximately 82% (n = 23) were female. Participants were composed of eleven presidents and screening program managers, seven nurses, seven sonographers, and three radiologist from nine MCH hospitals in Ordos City.

CFIR constructs were rated and influencing factors were summarized in Table 2. Six constructs were predominantly conductive to the implementation performance. Two constructs exerted negative influences, where the impact of barriers outweighed that of facilitators. One construct displayed neutral influence, where the impact of barriers and facilitators was comparable. Details are described below. All constructs except policies & laws and implementation leads constructs received mixed comments.

Table 2.

CFIR construct rating of organized breast cancer screening program in Ordos

Domains	Constructs	Ratings	Facilitators	Barriers
Outer setting	Policies & laws	+ 2	• Political commitment to cervical cancer elimination	-
	Partnerships & connections	+ 2*	• Multisectoral collaboration	-
	Performance-measurement pressure	+ 1*	• Rational screening goal-setting	• Migration
Inner setting	Available resources	0*	• Ultrasound machines generally available • Half staff with adequate work experience and educational level	• Mammography not generally available • Sonographers lacking breast-related knowledge • No breast surgeons
	Access to knowledge & information	−1*	• Regular training • Wide information coverage	• Limited professional training opportunities • Lack of content depth and timely updates
Individuals	Implementation leads	+ 2	• Specialized leadership	-
Process	Teaming	+ 2*	• Structured screening teams	-
	Engaging	+ 1*	• Various engaging strategies • Utilization of county-township-village networks	• Difficulties in preparing brochures • Low awareness of target population • Culture difference
	Doing	−1*	• Clearly articulated work plan	• Inconsistent referral • Ambiguous feedback of detected abnormalities • Untimely data entry • Non-interoperable electronic systems

^*added if comments were mixed

Outer setting

Policies & laws

There was favorable policy environment for two-cancer screenings in China. In line with the Cervical Cancer Elimination Initiative launched by the World Health Organization (WHO) on November 17, 2020, the NHC issued the Action Plan for Accelerated Elimination of Cervical Cancer (2023–2030). It was proposed that by 2025, the cervical cancer screening rate would reach 50%, and by 2030, it would reach 70%. Moreover, Chinese government has launched two-cancer screenings for women aged 35–64 and introduced this program in the national Public health services since 2009. The Government’s focus and commitment to eliminating cervical cancer has contributed significantly to the introduction and extension of the BCS. Explicit guidelines, clearly defined assessment indicators, and specific funds for the BCS were granted. Therefore, political commitment to cervical cancer elimination was recognized as a strong facilitator of the BCS implementation performance.

Partnerships & connections

Multisectoral involvement was identified as a facilitator. Figure 1 displays how MCH hospitals were networked with external health and non-health sectors. Health education was a necessary step before screening to motivate target population. MCH screening teams, the local Women’s Federation, and media were responsible for health education. Enterprises, schools, and communities helped notify and organize specific female groups for BCS. Women were mainly organized by their work units. Special groups like freelance women, non-occupational women, and female farmers/herders were organized by community, local Women’s Federation, or local labor union.

Fig. 1.

Partnership map of MCH hospitals with external health or non-health sectors

“The participation of local Women’s Federation in health education releases our work pressure a lot.” (P7)

“Our banner is vast but sparsely populated. The number of ethnic minorities and farmers/herders is large and they are difficult to engage in screening. Minorities, freelancers, nonoccupational women, and female farmers/herders are organized by their community or local labor union. In this way, we can cover most target population to receive BCS in the region.” (P14)

As primary screening sites, MCH hospitals delivered screening services to target population and accommodated medical suggestions. Any suspected case was recommended to refer to higher level hospitals for diagnosis. The Municipal Health Commission or Municipal MCH hospital supervised the screening program and processed quality control annually. Academic institutions provided technical support to achieve program sustainability. MCH hospitals were also linked to Public Security Bureau, Health Insurance Bureau, and Census Bureau for data retrieval. These data were used to master the distribution of target population and calculate the screening assessment indicators.

“We closely connect with the Municipal Health Commission, Health Insurance Bureau, Public Security Bureau, etc. to obtain statistics for the calculation of annual assessment indicators. Some data is inaccessible due to the personal identifiability. So we need to communicate with them about what we want.” (P22)

Performance-measurement pressure

Rational screening goal-setting facilitated the screening performance. Cervical and breast cancer screening services were typically delivered concurrently. To simplify the calculation and implementation, the two-cancer screenings shared the same annual screening goals in Ordos City. Taking into account the population of local women aged 35–64, the municipal government and the Municipal Health Commission set the annual screening goals as shown in Table 3. From 2021 to 2023, the annual screening completion rate in Ordos City was above 80%. The year 2023 performed the highest completion rate. However, some participants argued about migration issue. Out-migrant group caused overestimated screening target number.

Table 3.

Completion of annual breast cancer screening goals in Ordos

	Total female population (per 10,000)*	Female population aged 35–64 (per 10,000)*	Screening goal (person/year)*	Year	Actual screening number (person/year)	Completion rate (%)
Ordos City	89.4	40.5	48,600	2021	39,166	80.59
				2022	41,226	84.83
				2023	59,488	122.40
Kangbashi District	2.8	1.2	1440	2021	1472	102.22
				2022	1633	113.40
				2023	2177	151.18
Dongsheng District	18.8	8.5	10,200	2021	7633	74.83
				2022	6675	65.44
				2023	11,357	111.34
Dalad Banner	18.1	8.3	9960	2021	9969	100.09
				2022	9506	95.44
				2023	12,170	122.19
Junggar Banner	16.2	7.0	8400	2021	7031	83.70
				2022	7700	91.67
				2023	9354	111.36
Otog Front Banner	4.0	1.9	2280	2021	1831	80.31
				2022	2286	100.26
				2023	3188	139.82
Otog Banner	5.7	2.7	3240	2021	2967	91.57
				2022	3147	97.13
				2023	2906	89.69
Hangjin Banner	8.1	3.8	4560	2021	2031	44.54
				2022	2537	55.64
				2023	3670	80.48
Uxin Banner	5.8	2.6	3120	2021	2363	75.74
				2022	3018	96.73
				2023	4368	140.00
Ejin Horo Banner	9.9	4.5	5400	2021	3869	71.65
				2022	4724	87.48
				2023	10,298	190.70

^*Data from “Ordos City New Cervical and Breast Cancer Screening Technology Plan (2022 Edition)”

“The annual screening goals set by the government are calculated on the basis of the household population. However, there is a large number of household females moving out for work, resulting in an overestimated screening target number. The goal is hard for us to accomplish.” (FG2)

Inner setting

Available resources

Available resources in this study contained both personnel and equipment. Its overall impact on the screening performance was neutral. Qualified screening staff and generally availability of ultrasound machines were examined as facilitators. Not fully covered mammography services, limited knowledge of sonographers on breast cancer, and no breast surgeons were considered barriers.

As shown in Table 4, there were a total of 30 health workers who were able to undergo examination, 21 for BUS examination, 6 for Mammography, and 3 for CBE only. More than half were majoring in medical imaging (53.33%, n = 16), had a bachelor’s degree or higher (53.33%, n = 16), and were associate chief physicians or attending doctors (70%, n = 21). Around 43% (n = 13) had worked more than 10 years. Each MCH hospital owned at least one ultrasound machine. Two banners had no mammography machine. Most equipment was self-funded (64.00%, n = 16), and was Purchased after 2016 (72.00%, n = 18).

Table 4.

Available resources of organized breast cancer screening program in Ordos

Personnel	N (%)	Equipment	N (%)
CBE	24 (80.00)	BUS	18
BUS	21 (70.00)	MAM	7
MAM	6 (20.00)	BUS Manufacturer
Major		Samsung	2 (11.11)
Clinical medicine	9 (30.00)	GE healthcare	3 (16.67)
Medical imaging	16 (53.33)	Mindray	4 (22.22)
Nursing	2 (6.67)	SonoScape	1 (5.56)
Others	3 (10.00)	Philips	8 (4.44)
Professional title		MAM Manufacturer
Chief physician	2 (6.67)	SINO MDT	5 (71.42)
Associate chief physician	13 (43.33)	United Imaging	1 (14.29)
Attending doctor	8 (26.67)	Hologic	1 (14.29)
Others	7 (23.33)	Source of funding
Education level		Government-funded	2 (8.00)
Associate degree	14 (46.67)	Research projects	7 (28.00)
Bachelor’s degree or higher	16 (53.33)	Self-funded	16 (64.00)
Work experience		Year of purchase
< 5 years	9 (30.00)	Before 2016	7 (28.00)
5–10 years	8 (26.67)	After 2016	18 (72.00)
> 10 years	13 (43.33)
	N = 30

CBE clinical breast examination, BCS breast cancer screening, MCH maternal and child healthcare, CBE clinical breast examination, BUS breast ultrasound, MAM mammography

There were no breast surgeons in any of the nine MCH hospitals. Most staff participating in the organized BCS were sonographers and nurses. Many mentioned that their main business was ultrasound prenatal examination and they comprehended little breast expertise. Moreover, all mammography machines were Purchased after 2019. Radiologists were admittedly inexperienced in MAM operation and image interpretation.

Access to knowledge & information

The training opportunities of screening staff were limited and appeared to negatively impact the screening performance. Regular training and wide information coverage were potential facilitators. Limited professional training opportunities and lack of content depth and timely updates were potential barriers. Screening staff were trained at least one time annually. There were management training and professional training, either online or offline. Program management training was typically the interpretation of work plans provided by the Municipal Health Commission or Municipal MCH hospital. Professional training was mainly theoretical lectures, including epidemiology, clinical examination methods, Breast Imaging-reporting and Data System (BI-RADS) report standards, etc. It was delivered by higher level hospitals or academies. Several sonographers stated their lack of breast-related knowledge and clinical skills, and expressed their desire for more on-site practices.

“We do not specialize in breast diseases. When it is hard to report BI-RADS categories, we have no one for advice as our hospital has no breast experts. Our radiologists are also not experienced in mammography. I am looking forwards to more training opportunities in cancer hospitals and meeting with more cancer patients.” (P2)

Individuals

Implementation leads

Specified leadership appeared to be a facilitator. MCH hospitals in Ordos City all designated persons to manage the organized BCS program. Their familiarity with the BCS significance and their personal dedication promoted the smooth running of this program. During the interviews, all leaders demonstrated their thorough understanding of the screening work plan. Many of them had endeavored to optimize the procedure and enlarge the screening coverage.

“Our hospital has no MAM. BUS-positive women are recommended to appoint MAM as a supplement. But its recall rate is low. We therefore consulted with local People’s Hospital and obtained their permission to build referral channel. Suspected cases were referred to local People’s Hospital to supplement MAM and even biopsy there. This helped to increase our recall rate.” (P16)

Process

Teaming

Screening teams in all MCH hospitals shared explicit structures (Fig. 2), and structured screening teams was considered a facilitator. Directors of the Department of Gynecology were frequently appointed as leaders of two-screening teams. Each screening team was consist of gynecologists, sonographers, and nurses. They exhibited skilled linkage and demonstrated clear understanding of their job description. Each screening team performed moderate size (6–8 persons) and served approximately 100 females per day. They would adjust their team size accordingly.

Fig. 2.

Structure of a screening working group. The screening workflow contained several steps, such as engagement, health education, registration (including informed consent), survey, examination, information entry (including follow-up information), and quality control. Engagement such as call reminders were provided before women came to screen. Nurses guided the brief health education, registration, and survey. The survey before examination collected information such as history of breast diseases, family history of cancers, maternal history, menstrual conditions, and lifestyle to roughly evaluate the exposure to breast cancer risk factors. Gynecological examination and sampling of the HPV test were performed by two gynecologists. BUS examination was conducted by two sonographers (one for examination and one for report writing). Information entry into the electronic system were performed also by nurses. At the end, team leaders or one of the nurses took the obligation in quality control of the whole process

“The current team size is reasonable, neither too small to undermine service quality, nor too big to waste human resources.” (P17)

Engaging

There were mixed comments in target population engagement. Facilitators involved various engaging methods and utilization of county-township-village networks. Barriers included difficulties in preparing brochures, low awareness of target population, and culture difference. However, engagement is extensively recognized as a positive influencing factor.

Figure 3 illustrates six main methods that MCH hospitals used to engage target population: health education, family member engagement, gift incentives, countryside screening, group notification, and reminders. The local Women’s Federation, MCH staff, and media in Ordos City guided the health education by social media outreach, TV propaganda, lectures, festival activities on Women’s Day, and distribution of brochures. Diverse communication channels boosted the information dissemination. Gifts such as umbrellas and aprons were presented as incentives. Family members were also educated to promote screening uptake. Females not screened within three years would receive regular reminders by phone calls or messages. Community, enterprises, schools, local labor union, etc. were liable to collectively organize different groups of women to screen. This helped to reach the most of target population. For women living in remote rural areas, the county-township-village network was leveraged to increase accessibility. MCH hospital staff approached township health centers to provide screening services, and village doctors notified women to come.

Fig. 3.

Strategies to facilitate the screening uptake of target population

“Village doctors are working on the frontline to safeguard people’s health. By fully utilizing county-township-village network, we can bring BCS service to women’s side.” (P28)

However, problems were also noted during interviews. Some participants mentioned their difficulties in preparing brochures due to their limited time and experience. Moreover, participants generally agreed that target population were low in health awareness, and this intensively impeded the engagement. Ordos City had a large minority population. Language and cultural differences led to communication difficulties.

“We are not experienced in preparing of two-cancer screening brochures for dissemination and it is time-consuming. Thid work has crowded out our time for other businesses. We are now under pressure.”(P15)

“Women’s health literacy needs urgent improvement. Those who want to be screened come year after year, but those who do not never come, no matter how hard we try to persuade them.”(P3)

Doing

There were also mixed opinions on the execution of the organized BCS program. Clearly articulated work plan was regarded as a facilitator. Inconsistent referral, ambiguous feedback of detected abnormalities, untimely data entry, and non-interoperable electronic database were denoted as barriers. Considering that the negative comments outweighed the positive comments, the execution of this program was suggested as a negative influencing factor.

The organized BCS program has been introduced in Ordos City for more than ten years. Its workflow was complete and legible, as shown in Fig. 4. The screening process was practically divided into four sections, registration, examination, follow-up, and data management. Its instruction described the management of detected abnormalities clearly. However, some participants mentioned that improvements were needed in follow-up and data quality. Inconsistent referral from imaging to biopsy contributed to low compliance and difficulties in follow-up. Additionally, the detected abnormalities often failed to explain their situations or tended to conceal their results of following examinations. Some of their follow-up information was missed. Data was principally entered in real time. But paper questionnaires were adopted in case of poor network or system maintenance. There was then a lag in data entry, ranging from one week to six months. Several participants also raised the issues of the current electronic systems. The BCS system and the MCH hospitals’ Hospital Information System (HIS) were not interoperable. This created an information barrier between organized screening and opportunistic screening.

Fig. 4.

Breast cancer screening workflow in Ordos. Target population voluntarily signed the informed consent form, registered, and completed a survey. They then received BUS examination as the primary screening. Detected abnormalities would supplement Mammography. Mammogram-positive cases would be recommended for pathological biopsy and treatment accordingly. Their results were obtained by follow-up. BUS- or mammogram-negative ones would be suggested to attend screening every 2–3 years. Records including personal information, results of BUS examination, mammography, biopsy, and treatment were entered into the electronic system

“The BCS system and our HIS system are independent. In addition to this BCS program, we also provide with opportunistic breast cancer screening and other ultrasound examinations. Frequent switching between systems complicates our work procedures.”(FG1)

Discussion

This study evaluated the implementation of the organized BCS program in Ordos City and recognized its potential influencing factors from MCH hospitals’ perspectives. We conducted this research in the context of the national two-cancer screenings for women aged 35-64 [19, 20].

Outer setting

The outer setting CFIR domain illustrated the importance of political commitment to cervical cancer elimination and multisectoral collaboration. Policy support predominantly creates an enabling environment. Two-cancer screening program established mechanisms and provided guidelines to maintain efficiency and sustainability in BCS service delivery. Studies have reported that integrating BCS with cervical cancer screening enhanced reach and feasibility of both, and promote the early detection of breast cancer [32, 33]. Our study described the same finding.

WHO Cervical Cancer Elimination Initiative proposed that each country should meet the 90-70-90 targets by 2030, which are 90% of girls fully vaccinated with the HPV vaccine by the age of 15; 70% of women screened using a high-performance test by the age of 35, and again by the age of 45; 90% of women with pre-cancer treated and 90% of women with invasive cancer managed [34]. Chinese government stands with WHO Cervical Cancer Elimination Initiative, and adopted 90-70-90 targets in two-cancer policy. Local governments, like Ordos City, calculated the annual screening goal based on these targets. Cervical and breast cancer shared the same screening goals.

Apart from WHO Cervical Cancer Elimination Initiative, WHO also established Global Breast Cancer Initiative (GBCI) in 2021, with the hope to bring together stakeholders and reduce breast cancer [35]. It employs three key strategies, health promotion and early detection, timely diagnosis, and comprehensive breast cancer management. Unlike political commitment to cervical cancer elimination, corresponding targets have not been adopted in Chinese governmental documents. WHO Cervical Cancer Elimination Initiative has greater impact on the breast cancer screening in China, compared to GBCI.

The establishment of partnerships with other health and non-health sectors profoundly achieved an effective division of obligation. It is helpful to reduce MCH hospitals’ pressure. Similar publications also document that multisectoral collaboration leverages limited resources and increases the efficiency of health delivery systems [36, 37]. The screening completion rates in 2021, 2022, and 2023 were high. It proportionately promoted the motivation of screening staff so that they would not treat the BCS work as a burden. The completion rates in 2021 and 2022 were relatively lower. This may due to the strict restriction of indoor activities during COVID-19.

Migration was identified as a barrier. Ordos City has a large number of out-migrant workers. They were regarded as part of the target population, but their needs might have been overlooked. It is advisable to inform and encourage this special group to undergo local screenings in their working cities. MCH hospitals should also follow up the results.

Inner setting

The inner setting CFIR domain pointed out the importance of capacity building. Frontline health workers are essential to safeguard people’s health. Studies have reported that building the capacity of primary health workers contributes to cancer prevention and early detection in a low-cost way [38–40]. Moreover, the accuracy of ultrasound examination relies prominently on sonographers’ competence. Improving their capacity building is conducive to improving BCS service quality. However, most primary health workers have limited educational attainment or knowledge globally [41–44]which is unfavorable to health service delivery. Breast cancer imaging infrastructures in primary health care are deficient as well [45, 46] aligning with our findings. No mammography, pathological biopsy, or surgical supply in MCH hospitals attenuated the convenience of supplemental examination and diagnosis, resulting in a low recall rate of detected abnormalities.

Individuals

The individuals CFIR domain recognized specialized leadership as a facilitator. Two-cancer screening program managers in all MCH hospitals asserted strong leadership to advocate external collaboration, optimize the screening process, and promote screening coverage. According to Chatterjee R [47] leaders can act as catalysts to remain successful in health care delivery. A team of nursing leaders in a federally qualified health center in the US once collaborated and developed innovative strategies to successfully translate BCS guidelines into clinical practice and improve its coverage rate [48]. This highlights the crucial role of leader commitment in health intervention.

Process

The process CFIR domain could be ordinarily reorganized into four modules: health education, care delivery, follow-up, and data management. The Chinese government has documented improving people’s health literacy as one of the most fundamental, economic, and effective measures to improve the health level of the entire population [49]. We found that engaging strategies were tailored in accordance with the local context in Ordos City. The county-township-village network was judiciously leveraged to fully utilize resources of different levels for equitable allocation. It was first demonstrated in Dingxian County [50] and has been retained in the modern primary health care system in China [51]. Several studies have mentioned its prominent role in health education [52, 53].

The quality of care delivery highly relies on the medical team structure and health care instructions. Babiker [54] concluded that an effective teamwork can positively affect patient safety and outcomes. Our study also found that the structured screening teams and clearly articulated work plan achieved a clear understanding of job description and increased efficiency.

Inconsistent referral was a major consideration in our study to hamper the screening performance. This was reported to attenuate the adherence of target population to following imaging or biopsy, and exacerbate diagnostic delays [55] which is align with our findings. Not prompt referral, as well as ambiguous feedback of detected abnormalities, elicited the loss to follow-up. Poor data management due to untimely data entry and non-interoperable electronic database negatively affected the data quality and may cause bias in reflecting the real-world conditions.

Ordos City is taking the lead in two-cancer screenings in China. Its 5-year screening coverage rate of women aged 35–64 from 2016 to 2020 was higher than the national level during 2018-2019 [21, 22]. All facilitators discovered in this study contribute to it. However, this coverage rate is still way below that of other developed countries, such as Sweden (90%) and the US(80%) [56, 57]. Moreover, The BUS-positive rate in Ordos City was high, but the overall breast cancer detection rate was lower than the national average level in China [9, 21]. Identified barriers could be attributable to this problem. Underqualified professional competence and inconsistent referrals may reduce the adherence of the suspected cases to supplemental testing, resulting in a delay in diagnosis.

GBCI identified potential causes of deficits in breast-health systems and also found that lack of training, delays in biopsy, and delays in diagnosis were unfavorable for the performance of breast screening [58]. Mammography machine and breast surgeon deficiency is supposed to hinder follow-up, diagnosis and treatment, affecting the screening continuum. We did not investigate these resources at secondary or tertiary care facilities beyond the MCH. However, as far as we know, these resources are available at higher level care facilities. The diagnosis of breast cancer is also achieved in these hospitals. With the aim to better allocate the health resources by hierarchical care, we put more emphasis on the importance of establishing a sound referral mechanism.

Lessons learned from the implementation of the organized BCS program in Ordos City include (1) advocating for policy support; (2) bringing together multiple sectors, both health and non-health ones; (3) encouraging leadership input; and (4) building structured teams and articulated work plans. The needs for improvement in Ordos City include (1) strengthening the establishment of referral systems; (2) promoting the capacity building of primary health workers; and (3) enhancing the interactivity of electronic systems and information literacy of screening staff.

The main strength of this study includes (1) providing insights into the implementation of BCS program in Ordos City from caregivers’ perspectives, and (2) generating in-depth understanding of its influencing factors by using qualitative methodology. There are potential limitations. Firstly, there is a lack of views from other stakeholders, such as policy makers and care recipients (target population). Secondly, interview participants subjectively believed that target population were low in health awareness, but we did not investigate individual awareness level on BCS. Quantitative studies are necessary to investigate the knowledge and attitudes of recipients toward BCS. Lastly, this study might overlooked the implementation needs. Some of identified barriers are hard to be improved in the setting.

Conclusions

The implementation of the organized BCS program in Ordos City went overall well from caregivers’ perspectives. Political commitment to cervical cancer elimination, multisectoral collaboration, specialized leadership, and structured screening teams are four main positive drivers contributing to the implementation of the organized BCS in Ordos City. All facilitators contribute to the relatively high screening coverage rate. Inconsistent referrals, underqualified professional competence of primary health workers, and non-interoperable electronic systems are the main barriers. They are supposed to affect the screening quality and cause difficulties in follow-up management. We recommend that Ordos City facilitates the prompt referral for imaging and biopsy, promotes the capacity building of primary health workers, and strengthens the construction of electronic systems.

Abbreviations

NHC

National Health Commission

BUS

Breast ultrasound

CBE

Clinical breast examination

HPV

Human papillomavirus

MCH

Maternal and child healthcare

BCS

Breast cancer screening

CFIR

Consolidated Framework for Implementation Research

FG

Focus group

Authors’ contributions

Z.C. designed the study. Z.C., S.W., and H.Y. generated the interview outline and conducted interviews. N.C. and Y.Z. contacted study participants and provided information. Z.C. and H.D. coded the scripts and analyzed the data. Z.C. wrote the first draft. H.D. revised the manuscript. Y.Q. and C.W. provided advice and guidance. All the authors read and approved the final manuscript.

Funding

This study was supported by the Chongqing Tencent Sustainable Development Foundation through the project “Comprehensive Prevention and Control Demonstration Project for Eliminating Cervical Cancer and Breast Cancer in Low Health Resource Areas of China” (Project No. SD20240904145730).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

This study was approved by the research ethics committee of the Chinese Academy of Medical Sciences and Peking Union Medical College (approval number CAMS&PUMC-IEC-2022-059). Participation in this study was voluntary and we received informed consent from all the participants. All methods were carried out in accordance with the Declaration of Helsinki.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.