Effects of a telephone intervention on breast, cervical, and colorectal cancer screening: A meta-analysis.

1
Introduction
Cancer has been a major cause of death worldwide, with around 20 million new cases and around 10 million deaths in 2022; and there were regional differences, with Asia having a higher incidence and mortality than the Americas and Europe (Bray et al., 2024). Patients with early stage cancer have higher survival rates than do patients with late stage cancer (Araghi et al., 2021). Cancer screening has various benefits, including early detection, reduction in incidence and mortality rates, and better treatment options for breast (Monticciolo et al., 2021), cervical (Bouvard et al., 2021), and colorectal cancers (Knudsen et al., 2021).
Undergoing cancer screening involves knowledge, perceptions of the necessity of screening, overcoming barriers, and then making an appointment or taking action (Duffy et al., 2017). Various interventions or strategies have been employed to enhance cancer screening, including those targeting clients or providers, mass or small media, group or individual education, and reminders or health education (Brouwers et al., 2011; Duffy et al., 2017).
Many countries have a policy for cancer screening. People who meet the criteria for cancer screening can be contacted by telephone, through which they can be reminded of their screening schedule or educated on health issues. Previous reviews have shown the positive effects of client reminder on cancer screening (Brouwers et al., 2011; Dougherty et al., 2018; Duffy et al., 2017; Gruner et al., 2021). The effects of telephone interventions on cancer screening have been inconsistent, with some reviews showing better effects than with usual care (Agide et al., 2018; Duffy et al., 2017; Rat et al., 2018; Tsipa et al., 2021) but others showing worse effects than with letter or email interventions (Gruner et al., 2021). In the clinical setting, cancer screening uptake occurs when people visit primary care practitioners for routine care. However, telephone intervention aims to promote people in the community to take action, shifting screening from a visit-based healthcare intervention to a population-based one.
However, some of these reviews were not meta-analyses (Agide et al., 2018; Duffy et al., 2017; Gruner et al., 2021; Rat et al., 2018). In fact, one review that compared letter interventions with control and telephone interventions with control separately did not directly compare telephone interventions and letter interventions (Gruner et al., 2021). The telephone and letter are common strategies for patient outreach; except for the comparison between the telephone and the control group, we also wanted to know whether the telephone had higher effects than the letter. This study took telephone communication as an interventional tool/strategy, and aimed to explore the effects of telephone interventions on cancer screening behavior, among: (1) cancers: breast, cervical, and colorectal cancer; (2) telephone strategies: counseling and reminder; (3) follow-up periods: six months below, seven to 12 months, and 13 months and above; and (4) telephone strategies and cancers: counseling or reminder on breast, cervical, and colorectal cancer. To explore the effects of telephone interventions, we compared (1) telephone interventions versus control, (2) telephone interventions versus letter interventions, and (3) a combination of telephone and letter interventions versus letter interventions alone.

2
Methods
2.1
Design
A meta-analysis was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement (Moher et al., 2009). Ethical approval is not required because the analysis was based exclusively on publicly available data.

2.2
Search terms
Four electronic databases, including PubMed (Medline), Web of Science, Embase, and the Cumulative Index of Nursing and Allied Health Literature (CINHAL), were searched from inception to June 2024. The following keywords and strategies were used for the search: (telephone or phone call or telephone reminder or telephone counseling) and (cancer screening or mammography or pap smear or Papanicolaou test or fecal occult blood test or hemoccult) and (effect or adherence or attendance or compliance or uptake) and (randomized controlled trial or randomized controlled trial or RCT).

2.3
Inclusion criteria and exclusion criteria
The inclusion criteria were as follows: (1) studies that included screening behaviors for breast, cervical, and colorectal cancers as the dependent variables; (2) those that involved a telephone intervention or a combination of telephone and letter intervention, compared to a control or letter intervention; (3) those that used a randomized controlled design; and (4) those published in English. The control group received routine care without telephone or letter intervention, or no intervention.
The exclusion criteria were as follows: (1) studies that did not use a randomized controlled designed, did not focus on the three cancers, and did not measure cancer screening behaviors as dependent variables; (2) interventions that involving not only telephone or letter but also other components, such as chart reminder, physician talk, mailed kit, or promotes; (3) those that did not provide original data about the numbers of behaviors and sample sizes; and (4) review articles and correlational studies.
This search strategy was employed in all four databases, utilizing titles and abstracts to identify studies that potentially met the inclusion criteria. The full texts of the potential studies were retrieved and assessed by the research team.

2.4
Quality appraisal
The first (WLH) and second authors (IML) assessed the quality of the studies based on the Revised Cochrane Risk of Bias tool for randomized trials (RoB 2.0) (Sterne et al., 2019) using the following six domains: bias arising from the randomization process, bias due to deviations from intended interventions, bias due to missing outcome data, bias in outcome measurements, bias in the selection of the reported results; and overall risk of bias over three levels: low, some concerns, and high. Disagreements between reviewers were resolved through consensus with the fourth author (SYF).

2.5
Data abstraction
A standardized form was developed to extract data, including country, authors, demographic characteristics of the participants (age, gender, inclusion and exclusion criteria), characteristics of the intervention (counseling or reminder, theory of counseling, follow-up periods), measurement outcomes (sample size of each group, number cancer screenings after the intervention), and risks of bias. Two authors (WLH and IML) extracted the data independently, and discrepancies were resolved through discussion with all authors.

2.6
Data synthesis
The number of cancer screenings after intervention and the sample size of each group were coded. Odd ratios were used to determine the effects of the interventions, which included telephone versus control, telephone versus letter, and both telephone and letter versus letter. The Chi-square test and I2 statistic were used to determine the heterogeneity between studies, with an I2 value over 50% indicating substantial heterogeneity. Given the heterogeneity of the included studies, a random-effects meta-analysis was used. In addition, subgroup analyses were also conducted due to high heterogeneity, including three cancers (breast, cervical, and colorectal cancer), two telephone interventions (counseling and reminder), three follow-up periods (six months below, seven to 12 months, and 13 months and above), and three cancer screenings × two telephone interventions. The I2 statistic was used to test the heterogeneity in subgroup analyses. Review Manager 5.4 was used for data analysis, and a p-value of less than 0.05 indicated statistical significance.

3
Results
3.1
Study selection
After removing duplicate articles, a total of 791 journal articles were identified from the four databases using the search strategies. After screening the titles and abstracts, the full texts of 129 articles were obtained. A total of 62 articles were excluded for not providing the frequency and sample size details (N = 17); not randomly assigning participants (N = 4); not focusing on the target cancers (N = 1, not breast, cervical, or colorectal cancer screening), interventions (N = 33, not telephone, letter, control, or mixed multiple components), or outcomes (N = 5, not cancer screening behavior), having no control group (N = 1); or having repeated data (N = 1). Five additional articles were included from manual searching of the reference lists, and ultimately, 72 articles were included for meta-analysis. Fig. 1 presents the flow chart of article searching, inclusion, and exclusion.
Screening methods included mammography for breast cancer (35 articles); pap test for cervical cancer (15 articles); and fecal occult blood test, sigmoidoscopy, or colonoscopy for colorectal cancer (33 articles). The countries included: Asia = 7 (Hong Kong = 1, Iran = 2, Israel = 1, Korea = 1, Malaysia = 1, Taiwan = 1), Europe = 13 (Belgium = 1, France = 2, Germany = 2, Italy = 1, Spain = 3, Sweden = 2, UK = 2), Oceania = 3 (Australia = 1, New Zealand = 2), North America = 48 (Canada = 3, USA = 45), and South America = 1 (Brazil = 1). Additionally, 31 (43.06%) articles were published prior to 2010.

3.2
Methodological quality of the included studies
Table 1 presents the risk of bias of the included studies. A total of eight studies (10.11%) had some concerns in the randomization process, whereas six (8.33%) had some concerns in intended interventions and missing outcome data.

3.3
Intervention characteristics
Two types of interventions were delivered via telephone. The first intervention involved reminding the participants to undergo cancer screening. In general, the reminder included the importance of cancer screening, the reason they received the telephone, and the opportunity to schedule an appointment.
The second intervention was counseling. Some studies did not state a specific theory of counseling but focused on the beliefs, benefits, barriers, and self-efficacy of cancer screening, providing relevant information and appointment schedules. The specific theories used to develop intervention content included the conflict model of decision-making, the health belief model, motivational interviewing, social cognitive theory/social learning theory, the theory of reasoned action, the transtheoretical model, and the precaution adoption process model.
Regarding the type of intervention, three studies used automated telephone, two had electronic health records; and all letter interventions were physical mail, no electronic mail. The references used in the meta-analysis are listed in the Supplementary file.

3.4
Effects of telephone versus control
Compared to the control intervention, the telephone intervention had higher effects in the screening of all three cancers (odds ratio [OR] = 2.05, p < 0.01), breast cancer screening (OR = 1.62, p < 0.01), cervical cancer screening (OR = 2.13, p < 0.01), and colorectal cancer screening (OR = 2.54, p < 0.01). All of the I2 were larger than 50%, with high heterogeneity. Both counseling (OR = 2.22, p < 0.01) and reminder (OR = 2.28, p < 0.01) had positive effects on screening of all three cancers, with the effects occurring at one to six months (OR = 2.12, p < 0.01), seven to 12 months (OR = 2.11, p < 0.01), and above 13 months (OR = 1.73, p < 0.01). Regarding the specific types of interventions on the screening of different cancers, positive effects were observed in counseling and reminder for breast cancer screening (OR = 1.56, p < 0.01; OR = 2.17, p < 0.01), cervical cancer screening (OR = 2.07, p < 0.01; OR = 2.20, p < 0.01), and colorectal cancer screening (OR = 2.60, p < 0.01; OR = 2.51, p < 0.01). Table 2 presents the comparisons between the effects of the telephone interventions and the control groups, as well as the subgroup analyses, and Fig. 2 presents the forest plot of the comparison.

3.5
Effects of telephone versus letter intervention
Compared to letter interventions, telephone interventions had higher effects in the screening of all three cancers (OR = 1.45, p < 0.01), breast cancer screening (OR = 1.36, p < 0.01), and colorectal cancer screening (OR = 1.75, p < 0.01). Both counseling (OR = 1.55, p < 0.01) and reminder (OR = 1.40, p < 0.01) had positive effects on the screening of all three cancers, with the effects occurring at one to six months (OR = 1.57, p < 0.01) and seven to 12 months (OR = 1.35, p = 0.01). Regarding the specific types of interventions on the screening of different cancers, positive effects were observed for counseling on colorectal cancer (OR = 2.38, p < 0.01); and reminders on breast cancer screening (OR = 1.47, p < 0.01), cervical cancer screening (OR = 1.43, p = 0.04), and colorectal cancer screening (OR = 1.26, p = 0.03). All comparisons exhibited high heterogeneity, with an I2 value above 50%. Table 2 presents the comparisons between the effects of the telephone interventions and the letter groups, as well as the subgroup analyses, and Fig. 3 presents the forest plot of the comparison.

3.6
Effects of telephone and letter interventions versus letter intervention
Compared to letter interventions, combining telephone and letter interventions had positive effects on the screening of all three types of cancers across different follow-up periods. Counseling and reminders on the screening of all three types of cancers were all significant, except for the reminder on cervical cancer screening, for which no study conducted the comparison. The I2 for all three types of cancer screening, breast cancer screening, colorectal cancer screening, counseling and reminder, one to six months, above 13 months, and breast cancer and counseling were higher than 50%. Table 2 presents the comparisons between the effects of the combination of telephone and letter interventions and the letter groups, as well as the subgroup analyses, and Fig. 4 presents the forest plot of the comparison.

4
Discussion
4.1
Key findings
The meta-analysis explored the effects of telephone intervention on cancer screening behaviors. The telephone intervention had positive effects on the screening of breast, cervical, and colorectal cancer across three follow-up periods, and both counseling and reminder strategies were effective, based on the comparisons between telephone versus control and between the combination of telephone and letter versus letter. In addition, the telephone intervention had higher effects than did the letter intervention on the screening of breast and colorectal cancer, and both reminder and counseling were effective.

4.2
Interpretation of findings
Similar to previous systematic reviews (Agide et al., 2018; Duffy et al., 2017; Rat et al., 2018) but unlike the review of Gruner et al., which was not a meta-analysis (Gruner et al., 2021), the current meta-analysis revealed that those who received the telephone intervention were 2.05 and 1.54 times more likely to undergo cancer screening than the control group and the letter group, respectively.
Telephone interventions provide a means of outreach to remote areas (Inadomi et al., 2021; Tsipa et al., 2021) and one-on-one individual interaction (Fuzzell et al., 2021; Popalis et al., 2022), which are important factors for promoting cancer screening. Telephone communication has multiple functions in cancer screening. First, telephone communication can help remind people that their screening is due or overdue (Acharya et al., 2021; Brouwers et al., 2011). Reminders have been found to be an effective strategy for breast, cervical, and colorectal cancer screening (Rodriguez-Gomez et al., 2020). Second, telephone communication can also deliver screening information for health education or counseling (Brouwers et al., 2011; Inadomi et al., 2021). Third, it may remove structural barriers, such as making appointments or sending screening kits, which can improve their action (Brouwers et al., 2011; Inadomi et al., 2021).
Our results showed that both reminder and counseling were effective but may function differently. Reminders aim to make people become aware of the necessity of cancer screening. The general components of reminders in the studies included the reason why the participants received the telephone and how to make an appointment. Some studies focused on beliefs or barriers, and provided the interactions to answer the participants' questions. Regarding counseling, the health belief model and transtheoretical model have been the most widely used theories (Acharya et al., 2021). Healthcare staff have spent a considerable amount of time providing knowledge for education, and discussing the needs for and barriers to cancer screening with people and then helping them take appropriate action. However, counseling may also have the reminder function which makes people aware the necessary of cancer screening.
Both counseling and reminder of telephone had higher effects on cancer screening than the letter intervention. The subgroup analysis showed that counseling had a positive effect on colorectal cancer, whereas reminder on all three cancers. Gender differences and aversion to screening may play a certain role. Breast and cervical cancer screenings are only for women who may be familiar with the two screenings, whereas colorectal cancer screening can be utilized by both men and women. In addition, the dislike for colorectal cancer screening was higher than for cervical cancer screening (Lo et al., 2013). Therefore, people need telephone counseling, which can provide rich information and assistance to facilitate colorectal cancer screening.
The current review found high heterogeneity among the included studies, which was similar to previous reviews (Dougherty et al., 2018; Tsipa et al., 2021). First, the characteristics of the participants varied considerably. Age and sex differences were observed between patients who underwent breast and cervical cancer screening and patients who underwent colorectal cancer screening. Some studies have focused on those who did not undergo regular cancer screening, whereas others focused on those who did not respond to reminders, as well as those with low socioeconomic status or minority groups. Second, different countries with various healthcare systems also contributed to the heterogeneity. For example, the US has the EPIC best practices advisory to send a portal message about cancer screening, that most telephone reminders and letters are now automatically sent electronically, rather than through the mail. Other countries may not have the system (Whitfield et al., 2025). The use of the EPIC best practices advisory is heterogeneous, so some physicians and patients may disregard the reminder. Third, the intervention comprised a variety of components, including different focuses, theories, and interpersonal interactions between participants and researchers.

4.3
Clinical implications
Regarding clinical implications, telephone interventions can be integrated into clinical practice for cancer screening. The electronic medical record system can remind healthcare professionals that certain people require cancer screening, after which telephone interventions can be used to remind and instruct them how to complete their cancer screening. For those who received reminders but failed to take action, telephone counseling can be utilized to reduce barriers and improve motivation.

4.4
Strengths and limitations
Using cancer screening behavior as the outcome, we investigated the effects of telephone interventions based on randomized controlled studies. Some limitations should be acknowledged. High heterogeneity was observed among the included studies despite subgroups analyses for different cancers, time periods, types of interventions, and types of intervention and cancers. However, other potential confounding factors related to cancer screening behavior were not included. The relevant studies on telephone interventions were included, although some were outdated. In addition, the effects of the multicomponent interventions, the letter interventions versus the control groups, were not examined in this review. Due to technological advancements, cell phones, email, and text messages are increasingly used in clinical care, and patients' adoption of these strategies has also evolved over time.
Future studies can shift from clinic-based to population-based approaches, and compare the effects between reminder and counseling, as well as between different theories, single or multiple strategies, to identify the significant components of telephone interventions. In addition, the effects of technology, including email, text messages, automated telephone calls, and electronic health record portal messages, can be tested. Furthermore, cost-utility studies can be used to explore the financial benefits of telephone intervention.

5
Conclusion
Telephone interventions had positive effects on cancer screening behaviors, including breast, cervical, and colorectal cancer screening, as well as short- and long-term follow-up periods, reminder and counseling, and reminder or counseling on the screening of all three cancers.

CRediT authorship contribution statement
Wei-Lun Huang: Writing – review & editing, Writing – original draft, Software, Resources, Project administration, Methodology, Formal analysis, Data curation, Conceptualization. I-Mei Lin: Writing – review & editing, Writing – original draft, Supervision, Software, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Wan-Ting Huang: Writing – review & editing, Writing – original draft, Project administration, Methodology, Investigation, Formal analysis, Data curation. Sheng-Yu Fan: Writing – review & editing, Writing – original draft, Supervision, Software, Resources, Methodology, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization.

Declaration of generative AI and AI-assisted technologies in the writing process
This research did not use any generative AI and AI-assisted technologies in the writing process.

Funding
This work was supported by the National Science and Technology Council, R.O.C. [grant numbers: MOST 111–2410-H-006-068].

Declaration of Competing Interest
The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Sheng-Yu Fan reports financial support was provided by National Science and Technology Council, R.O.C. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.