J Med Syst (2014) 38:118 DOI 10.1007/s10916-014-0118-x
MOBILE SYSTEMS
Use of Text-Message Reminders to Improve Participation in a Population-Based Breast Cancer Screening Program C. Vidal & M. Garcia & L. Benito & N. Milà & G. Binefa & V. Moreno
Received: 25 February 2014 / Accepted: 16 July 2014 # Springer Science+Business Media New York 2014
Abstract To analyze the effect of a cell text message reminder service on participation in a mammogram screening program in Catalonia, Spain. A quasi-experimental design was used with women aged 50 to 69 years who had been scheduled mammogram appointments in June or July 2011. Women were personally invited by letter to attend to the breast cancer screening program (n=12,786). Prior to the invitation, 3,719 (29.1 %) of them had provided their cell telephone number to the National Health Service. These women received a text message reminder 3 days before their scheduled appointment. Logistic regression models were used to analyze whether the text message reminder was associated with participation in screening. Cost-effectiveness of adding a text message reminder to the invitation letter was also analyzed. The overall rate of participation in breast cancer screening was 68.4 %. The participation rate was significantly higher in the text messaging group, with an age-adjusted OR of 1.56 (95 %CI: 1.43–1.70). A detailed analysis showed that the increase in participation related to the text message reminder was higher among women without previous screening who lived in areas where access to postal mail was limited (OR=2.85; 95 %CI: 2.31–3.53) compared to those who lived in areas of easier This article is part of the Topical Collection on Mobile Systems
C. Vidal : M. Garcia (*) : L. Benito : N. Milà : G. Binefa : V. Moreno Cancer Prevention and Control Program, Catalan Institute of Oncology, IDIBELL, Av. Gran Via 199-203, 08908 L’Hospitalet de Llobregat, Barcelona, Spain e-mail:
[email protected] L. Benito Department of Fundamental Care and Medical-Surgical Nursing, University of Barcelona, L’Hospitalet de Llobregat, Barcelona, Spain V. Moreno Department of Clinical Sciences, University of Barcelona, L’Hospitalet de Llobregat, Barcelona, Spain
postal mail access (OR=1.66; 95 %CI: 1.36–2.02). The invitation letter+text message reminder was a cost-effective strategy. Text message reminders are an efficient cost-effective approach to improve participation in difficult-to-reach populations, such as rural areas and newly developed suburbs. Keywords Breast cancer screening . Participation . Text messages . Reminders . Cost-effectiveness
Introduction Whether breast cancer screening does more harm than good has been debated extensively. The main questions are how large the benefit of screening is in terms of reduced breast cancer mortality and how substantial the harm is in terms of overdiagnosis. Several review studies recently carried out conclude that screening reduces breast cancer mortality but that some overdiagnosis occurs [1, 2]. Information should be made available in a transparent and objective way to women invited to screening so that they can make informed decisions. The effectiveness of population-based screening programs substantially depends on participation [3, 4]. Reductions in mortality seen in randomized trials can only be reproduced in the wider population if participation is adequate [5]. An undue focus on high participation rate can preclude an appropriate consideration of issues such us informed decision making and informed choice. A screening program need to avoid problems of inequity in health service provision and to achieve participation rates that lead to substantial reduction in mortality while ensuring that screenees are making informed choices and are aware of all the risks and benefits of participation [6]. In Catalonia (Spanish Region) breast cancer screening programs invite women aged 50 to 69 years for a biennial mammogram by sending a letter with an appointment [7]. Participation in the organized programs for breast cancer
118, Page 2 of 7
screening is now 65 % [8]. Even among women who decide to participate, absences occur due to forgetfulness or misunderstandings [9]. Failure to attend appointments is a major cause of inefficiency, leading to a misuse of resources and unnecessary increase in costs, and seriously hinders program planning [10–12]. Our breast cancer screening program in the southern Barcelona metropolitan area has a high rate of absences, which leads to requests for a new appointment. The program covers a heterogeneous territory with a densely populated urban area and small scattered, coastal areas of recent urbanization (Fig. 1). In these lower density areas, the mail is distributed with less frequency and post office boxes are frequently used, leading to delays in receiving appointments. Current standards of communication based on telephone contacts also lead to checking the postal mail less often. Communication technologies have been intensively introduced into everyday life. Email and cell phone have been established as the main method of communication for many people, displacing traditional postal mail and the landline
J Med Syst (2014) 38:118
phone [13]. Public health is not a stranger to this reality, and health systems are in transition to the web society [14–16]. Cell phone communication is widespread among the population. The National Statistics Institute reported that 95 % of Spanish households had a cell phone in 2012 [17], and this device is used by more than 70 % of people in all age groups. Text messages sent through the Short Messaging Service (SMS) allow communication with a large number of individuals at a relatively low cost and are delivered almost instantly [13]. This method is less intrusive than a phone call because it can be read whenever it is more suitable for the receiver [18]. In addition, SMS reaches people in isolated areas better than landline phones. The present study assessed the usefulness of sending an SMS text as an appointment reminder for the breast cancer screening program. We measured the impact of this approach on attendance at appointments (immediate participation) and performed a simple cost-effectiveness analysis. Finally, SMS was assessed as a tool for scheduling and managing appointments.
Fig. 1 Population density in the South Metropolitan Area of Barcelona (inhabitants per Km2)
J Med Syst (2014) 38:118
Methods We conducted a quasi-experimental study in the breast cancer screening target population of the southern Barcelona metropolitan area. This area includes 155,000 women who are invited for screening mammography every 2 years. Approximately 40 % of the women have not participated previously, and the usual attendance rate of this group is near 30 %. Women who have participated previously have a participation rate of more than 80 %. For the study population, 12,786 women with a scheduled appointment were selected between June 13, 2011, and July 12, 2011. According to the regular procedures of the organized cancer screening programs, all women were invited by ordinary postal letter. Women with a cell phone number previously registered in the population-based database from the National Health Service (RCA from its acronym in Spanish), also received an SMS reminder 3 days before the scheduled appointment (n=3,719; 29.1 %). When a woman reports any changes regarding her personal information to the RCA such as an address change or a register with a different general practitioner, additional contact information (i.e. cell phone number) is gathered. An outsourcing company (ALHORA Solutions) was hired for sending SMS. Women whose cell phone numbers were not available became the control group. The system allowed an SMS reply requesting an appointment change. In that case, an SMS with a new appointment was sent automatically, not allowing women to provide date or time preferences. In contrast, women who received only an invitation letter had to reschedule their appointments by phone and were able to agree on a convenient date and time for their new appointment. A woman was considered a participant if she attended the first scheduled appointment or an appointment before October 31, 2011. Data on participation (mammography attendance), age, screening unit, and previous screening were retrieved from the program database. We hypothesized that low participation in some of our screening units was associated with postal mail difficulties. In our territory, postal mail delivery has shown some difficulties in rural areas and newly developed suburbs. In the less populated areas or in areas with recent increases in population, postal mail service is not optimal. Consequently, we classified our screening units (n=11) according to the population growth between 2003 (when the program was fully operative) and 2010 and the percentage of towns with less than 15,000 inhabitants. With this information, the screening units were classified as easy or hard to reach by postal mail. The study protocol was approved by the Clinical Research Ethics Committee of the Bellvitge University Hospital (PR260/13) and all involved parties followed the ethical requirements set forth in the Spanish Organic Law on Protection of Personal Data (15/1999 of December 13).
Page 3 of 7, 118
Logistic regression models were used to analyze whether the SMS reminder was associated with participation in the screening program. Age-adjusted odds ratios (OR) and 95 % confidence intervals (CI) were estimated. Previous screening behavior and accessibility were considered potential confounders and modifiers of SMS efficiency. We performed stratified and multivariate analysis to control for potential confounders behavior, accessibility and age), because the 3,719 women with cell phone may have better knowledge about the screening program and they maybe are more prone to attend such screening program since they updated their contact information in the RCA database. Cost-effectiveness analysis We analyzed the cost-effectiveness of adding a SMS reminder to the invitation letter (strategy 2) compared to the invitation letter alone (strategy 1). Participation in the breast cancer screening (mammography attendance) was considered the unit of benefit. The analysis was performed from the perspective of the payer, and it was calculated for 100,000 women with 40 % having no previous screening. We only considered the direct costs at the time of analysis, including the screening mammogram with double reading (28.2€), administrative management of the invitation process (2.57€), and the invitation (letter: 0.51€ and SMS reminder: 0.11€). Though cell phone numbers were available for only 29.1 % of the women in the study population, a more recent update of the database records retrieved cell phone numbers for 40 % of the target population. We used last percentage for the costeffectiveness analysis. We estimated the cost per capita as the annual cost divided by the number of participants in each strategy; and the cost per extra participant as the additional cost divided by the number of new participants (benefit) arisen by the SMS. These estimations were performed considering overall screening, initial screening and subsequent screening. Sensitivity analysis was performed to assess the effect of the proportion of available cell phone numbers on the costeffectiveness analysis. Cost-effectiveness was calculated for three percentages: The lower range value considered was the proportion of cell phone numbers available in our study (30 %), actual proportion of cell phone numbers available in target population (40 %), and the upper range value was the proportion of cell phone availability in Spain reported by the National Statistics Institute (70 %) [17]. We also analyzed model robustness in regards to the distribution of women based on previous screening. We considered a range from 50 to 80 % to explore different scenarios. One similar to a newly implemented screening program with a higher proportion of women without previous screening (50 %) and another closely resembling the distribution of a
118, Page 4 of 7
J Med Syst (2014) 38:118
well-established program with a higher proportion of women who previously participated (80 %).
Results Breast cancer screening participation A total of 74.9 % of the women who received the SMS reminder and 65.0 % of the women who only received the invitation letter attended their appointments (OR=1.63, 95 % CI: 1.49–1.78). When modeling the participation (Table 1), the SMS reminder was statistically significant irrespectively of the previous screening behavior and accessibility. However, the effect of the SMS reminder was larger on the group without previous screening behavior and among women who lived in hard to reach areas (OR=2.85; 95 % CI 2.31–3.53). The proportion of rescheduled appointments in the group receiving the SMS reminder was 8.3 %, and among women invited only by letter was 7.0 % (OR=1.20, 95 % CI 1.04– 1.38). After rescheduling an appointment, 74.2 % of the women who received the SMS reminder and 80.7 % of the women invited only by letter attended the new appointment (OR=0.69, 95 % CI 0.49–0.96).
the cost per capita was 46.42€ for the strategy ‘only letter’ and 44.93€ for the strategy ‘letter+SMS reminder’. In initial screening, the cost per capita was 1.49€ lower for the strategy ‘letter+SMS reminder’. On the contrary, it was 0.08€ higher in subsequent screening. The cost per extra participant with the strategy ‘letter+ SMS’ was 32.85€, 29.55€ and 38.33 € for overall screening, initial screening and subsequent screening, respectively. The sensitivity analysis performed regarding the availability of cell phone numbers are shown in Fig. 2. Among women without previous screening, the cost per capita was inversely proportional to the cell phone number availability for the strategy ‘invitation letter+SMS reminder’. The sensitivity analysis regarding the distribution of women with previous screening showed that strategy ‘letter+SMS’ was not cost-effective for screening programs with a higher proportion of previously screened population (>60 %). The cost per capita was lower (−0.04€) for the strategy ‘invitation letter+SMS reminder’ when the proportion of previous screening was 50 %. While increasing the proportion of previous screening, the SMS reminder strategy had the opposite effect, cost per capita differences ranged from +0.00€ with a 60 % of subsequent screening to +0.05 € with an 80 % of subsequent screening (data not shown).
Discussion Cost-effectiveness analysis The annual cost per 100,000 women invited to screening, cost per capita and the cost per extra participant for each strategy care considering 40 % of cell phones availability are shown in Table 2. In overall screening, there were no differences in cost per capita among both strategies (32.76€). In initial screening,
SMS reminders enhanced participation in our breast cancer screening program. This finding was in agreement with other authors [10, 19, 20], not only in cancer screening [21–26]. More women who received the SMS reminders attended their appointments compared to those who were only invited by letter. The reminder had less of an impact among women who
Table 1 Impact of the SMS reminders on the participation for breast cancer screening Previous screening behavior No
SMS reminder
Yes
n (%)
ORa
95 %CI
n (%)
ORa
95 %CI
1443 (35.3) 1437 (35.1)
1 0.85
(0.72–1,00)
2570 (30.7) 3435 (41.0)
1 0.84
(0.73–0.96)
671 (16.4) 542 (13.2)
1.66 2.85
(1,36–2,02) (2,31–3,53)
1038 (12.4) 1339 (16.0)
1.34 1.93
(1.07–1.67) (1.57–2.37)
No Accessibility Easy to reach areas Hard to reach areas Yes Accessibility Easy to reach areas Hard to reach areas
Reference category: women who lived in easy to reach areas and were invited to screening only by letter ORa: age-adjusted odds ratios were calculated by means of a Multivariate Logistic Regression Model stratified by previous screening behavior
J Med Syst (2014) 38:118
Page 5 of 7, 118
Table 2 Cost-effectiveness analysis based on previous screening behavior for breast cancer
Overall screening Initial screening Subsequent screening
Strategy
Participants
Annual cost
Cost per capita
Only letter Letter+SMS Only letter Letter+SMS Only letter Letter+SMS
56,017 57,345 8,537 9,366 47,480 47,979
1,834,937.00 € 1,878,563.00 € 396,317.00 € 420,816.00 € 1,438,620.00 € 1,457,746.00 €
32.76 € 32.76 € 46.42 € 44.93 € 30.30 € 30.38 €
Cost per extra participant
32.85 €* 29.55 € 38.33 €
Results per 100,000 women invited to screening and considering 40 % of cell phone number availability and 40 % of women without previous screening behavior * The marginal difference, masked by the rounding : 32.757€ vs 32.759 €, explains the higher cost per extra participant
had previously participated in screening compared to those who had not. This finding was not unexpected as adherence to the program exceeded 80 %, and the possibility of improvement in this group was small. Few studies have measured the impact of SMS reminders on population screening programs [21, 26, 27], and most of them have been conducted in healthcare units in which women knew in advance that they will be scheduled for an appointment [10, 28]. However, most of the studies have concluded that SMS is useful as a reminder tool. The main finding of this study is that SMS reminders increase participation in less populated areas or areas with
Fig. 2 Sensitivity analysis of the cost per capita of two breast cancer invitation screening strategies based on cell phone availability. * 1–2 refers to the invitation screening strategy (1: only letter, 2: invitation letter+SMS reminder); OS: overall screening; NPS: no previous screening; PS: previous screening
recent increases in population where postal mail service is not optimal. This effect is even larger among women who have not previously participated in the screening program. In that case, SMS reminders more than double the likelihood to participate. Another interesting result of this study was that SMS was not a useful tool for rescheduling appointments, as it did not boost the participation of women who requested rescheduling their appointment. This result was probably due to the automated SMS that did not offer the possibility of choosing a day or time. Our program currently offers a web-based system for changing appointments, and we are testing if this method
118, Page 6 of 7
improves participation for women who require a new appointment. Our simple cost-effectiveness analysis concluded that the overall improvement in participation was due to the benefit obtained in women without previous screening where the effect on participation was the highest with lower costs (−1.49 € for each participant). One limitation of the study is its quasi-experimental design. It could be argued that the 3,719 women with cell phone may have better knowledge about the screening program and they maybe are more prone to attend such screening program since they updated their contact information in the RCA database. However, we performed stratified and multivariate analysis to control for potential confounders (previous screening behavior, accessibility and age). Finally, the study was designed to estimate appointment attendance in the short-term (immediate participation), and the results should not be inferred to overall participation in the program (screening round participation). In today’s society, new technologies are displacing traditional communication systems. Screening programs are not immune to this change, and the present study demonstrates that adding an SMS reminder to the invitation letter is a cost-effective method of improving participation. In the case of regular participants, further research is needed to assess whether adherence to the program is or not affected by replacement of the letter by an SMS message. If not, the use of SMS would be even a more cost-effective alternative. We anticipate that the use of conventional mail for communication will be replaced by more efficient and secure alternatives, and testing initiatives such as SMS invitation to breast cancer screening programs is required. Some barriers need to be cleared before this approach can be fully implemented, mainly ensuring reception of the message, certification of the sender and receiver, and ensuring confidentiality of the personal information being sent [29–32].
Conclusion In summary, our study has shown that including a text SMS reminder increases participation in a cost-effective manner, especially for women participating for the first time in the breast cancer screening program and/or living in areas where ordinary postal mail has difficulties.
Acknowledgments The authors acknowledge Dr. Candela Calle and Dr. Jordi Trelis for their encouragement in carrying out this project work. This study was partially funded by the Carlos III Health Institute (RD/ 12/0036/0053, CIBERESP). LB is a PhD student in the Nursing Science Program at Barcelona University.
J Med Syst (2014) 38:118 Competing interests The authors declare that they have no competing interests.
References 1. Independent UK Panel on Breast Cancer Screening (2012). The benefits and harms of breast cancer screening: an independent review. Lancet 380:1778–86. 2. Summary of the evidence of breast cancer service screening outcomes in Europe and first estimate of the benefit and harm balance sheet (2012). EUROSCREEN Working Group J Med Screen 19 Suppl1:5–13 3. Bonfill X, Marzo M, Pladevall M et al. (2001) Strategies for increasing women participation in community breast cancer screening. Cochrane Database Syst Rev 1: CD002943. 4. Ascunce N, Salas D, Zubizarreta R, et al. (2010) Cancer screening in Spain. Ann Oncol. 21 (Suppl 3): iii43–51. 5. Barrat A, Mannes P, Irwig L, et al. (2002) Cancer screening. J Epidemiol Community Health 56:899–902. 6. Weller DP, Patnick J, McIntosh M Dietrich AJ (2009). Uptake in cancer screening programs. Lancet Oncol 10:693–9. 7. Ascunce N, Delfrade J, Salas D et al. (2013) [Breast cancer screening: characteristics and results of the Spanish programs]. Med Clin (Barc) 141:13–23. 8. Espinas JA, Aliste L, Fernandez E et al. (2011). Narrowing the equity gap: the impact of organized versus opportunistic cancer screening in Catalonia (Spain). J Med Screen 18:87–90. 9. Hasvold PE, Wootton R (2011) Use of telephone and SMS reminders to improve attendance at hospital appointments: a systematic review. J Telemed Telecare 17:358–64. 10. Car J, Gurol-Urganci I, de Jongh T et al. (2012) Mobile phone messaging reminders for attendance at healthcare appointments. Cochrane Database Syst Rev 7:CD007458. 11. Milne RG, Hornen M, Torsney B (2006) SMS reminders in the UK National Health Service: an evaluation of its impact on “no-shows” at hospital out-patient clinics. Health Care Manage Rev 31:130–6. 12. Cubillas JJ, Ramos MI, Feito RF, Ureña T (2012). An improvement in the appointment scheduling in primary health care centers using data mining. J Med Syst 38:89. 13. Atun RA, Sittampalam S (2006) A review of the characteristics and benefits of SMS in delivering healthcare. In: Atun RA, et al.editor(s). The role of cell phones in increasing accessibility and efficiency in healthcare.Vodafone Group PLC. 14. Free C, Phillips G, Felix L et al. (2010). The effectiveness of MHealth technologies for improving health and health services: a systematic review protocol. BMC Research Notes 3:250. 15. Riley WT, Rivera DE, Atienza AA et al. (2011) Health behavior models in the age of cell interventions: are out theories up to the taks? Transl Behav Med 1:53–71. 16. Krishna S, Boren SA, Balas EA (2009) Healthcare via mobile phones: a systematic review. Telemed J E Health 15:231–40. 17. National Statistics Institute. Spain in figures 2012. http://www.ine.es/ss/ Satellite?L=en_GB&c=INEPublicacion_C&cid=1259926334812&p= 1 2547 3511067 2&pagename=Produ ctosYServicios% 2FPYSLayout¶m1=PYSDetalleGratuitas (accessed 29 Sept 2012). 18. Atun RA, Gurol-Urganci I (2006) Analysis of calls to NHS Direct. In: Atun RA et al. editor(s).The Role of Mobile Phones in Increasing Accessibility and Efficiency in Healthcare. Vodafone Group PLC. 19. Guy R, Hocking J, Wand H et al. (2012) How effective are short message service reminder at increasing clinic attendance? A metaanalysis and systematic review. Health Serv Res 47:614–32.
J Med Syst (2014) 38:118 20. Vodopivec-Jamsek V, de Jongh T, Gurol-Urganci I et al. (2012). Mobile phone messaging for preventive health care. Cochrane Database Syst Rev 12: CD007457. 21. Gierish JM, DeFrank J, Bowling M et al. (2010) Finding the minimal intervention needed for sustained mammography adherence. Am J Prev Med 39:334–44. 22. Eilbert KW, Carroll, K, Peach J et al. (2009) Approaches to improving breast screening uptake: evidence and experience form Tower Hamlets. Br J Cancer 101: s64–s67 23. Chen ZW, Fang LZ, Chen LY et al. (2008). Comparison of an SMS text messaging and phone reminder to improve attendance at a health promotion center: a randomized controlled trial. J Zhejiang Univ Sci B 9:34–8. 24. Leong KC, Chen WS, Leong KW et al. (2006) The use of text messaging to improve attendance in primary care: a randomized controlled trial. Fam Pract 23:699–705. 25. Guy R, Wand H, Knight V et al. (2013) SMS reminders improve rescreening in women and heterosexual men with chlamydia infection at Sydney Sexual Health Centre: a before-and-after study. Sex Transm Infect 89:11–5.
Page 7 of 7, 118 26. Hanauer DA, Wentzell K, Laffel N et al. (2009) Computerized Automated Reminder Diabetes System (CARDS): e-mail and SMS cell phone text messaging reminders to support diabetes management. Diabetes Technol Ther 11:99–106. 27. DeFrank JT, Rimer BK, Gierisch JM et al. (2009) Impact of mailed and automated telephone reminders on receipt of repeat mammograms: a randomized controlled trial. Am J Prev Med 6:459–67. 28. Goel A, Geroge J, Burack RC (2008) Telephone reminders increase re-screening in a county breast screening program. J Health Care Poor Underserved 19:512–21. 29. Kuory E, Faris C (2005) Mobile phones and clinic appointments, the start of a beautiful new Friendship? Br J Health Computing & Information Management 22:18–20. 30. Chaudry R, Scheitel SM, McMurtry EK et al. (2007) Web-based proactive system to improve breast cancer screening: a randomized controlled trial. Arch Intern Med 167:606–11. 31. Car J, Sheikh A (2004) Email consultations in health care: 1-scope and effectiveness. BMJ 329:435–8 32. Car J, Sheikh A (2004) Email consultations in health care: 2. acceptability and safe application. BMJ 329:439–42.
