Journal of Adolescent Health 56 (2015) S17eS20
www.jahonline.org Original article
Effectiveness of Centralized Text Message Reminders on Human Papillomavirus Immunization Coverage for Publicly Insured Adolescents Cynthia M. Rand, M.D., M.P.H. a, *, Howard Brill, Ph.D. b, Christina Albertin, M.P.H. a, Sharon G. Humiston, M.D., M.P.H. c, Stanley Schaffer, M.D., M.S. a, Laura P. Shone, Dr.P.H., M.S.W. d, Aaron K. Blumkin, M.S. a, and Peter G. Szilagyi, M.D., M.P.H. e a
Department of Pediatrics, University of Rochester School of Medicine and Dentistry, Rochester, New York Monroe Plan for Medical Care, Rochester, New York Department of Pediatrics, Children’s Mercy Hospitals and Clinics, Kansas City, Missouri d Department of Research, Division of Primary Care Research, American Academy of Pediatrics, Elk Grove Village, Illinois e Mattel Children’s Hospital UCLA, Los Angeles, California b c
Article history: Received July 11, 2014; Accepted October 14, 2014 Keywords: Human papillomavirus vaccines; Reminder systems; Text messaging; Adolescent health services
A B S T R A C T
Purpose: We evaluated a managed care organization (MCO)egenerated text message remindere recall system designed to improve human papillomavirus (HPV) vaccination coverage. Methods: We conducted a randomized controlled trial of text remindererecall for parents of 3,812 publicly insured adolescents aged 11e16 years with no prior HPV vaccinations who were enrolled in a single MCO and were patients at one of 39 primary care practices. We determined the rate of HPV receipt for intervention versus control with the KaplaneMeier failure function and determined hazard ratios using a clustered stratiﬁed Cox model, clustering on primary care provider and stratiﬁed on practice. We examined results for all subjects, and for those with a valid phone number, stratiﬁed by age group (11e13 years and 14e16 years) and gender. A post hoc analysis included all subjects and controlled for age and gender. Results: HPV dose 1 vaccination rates were not signiﬁcantly different when all participants were included, but for the subset of parents (54%) able to receive messages, HPV dose 1 rates were 13% for the control group and 16% for the intervention group; hazard ratio, 1.3 (95% conﬁdence interval, 1.0 e1.6; p ¼ .04), when controlling for age and gender. There were no signiﬁcant ﬁndings in the analysis stratiﬁed by age and gender. Conclusions: MCO-based text reminders are feasible and have a modest effect on HPV dose 1 vaccination rates for those parents able to receive text messages with valid phone numbers in the MCO database. Future studies should examine a similar intervention for those parents who already accepted the ﬁrst HPV vaccine dose. Ó 2015 Society for Adolescent Health and Medicine. All rights reserved.
Conﬂicts of Interest: S.G.H. is a consultant to the not-for-proﬁt, Immunization Action Coalition. None of the other authors reports any conﬂicts of interest. Disclaimer: Publication of this article was supported by the Society for Adolescent Health and Medicine through a grant from Merck & Co., Inc. The opinions or views expressed in this supplement are those of the authors and do not represent the position of the funder. 1054-139X/Ó 2015 Society for Adolescent Health and Medicine. All rights reserved. http://dx.doi.org/10.1016/j.jadohealth.2014.10.273
IMPLICATIONS AND CONTRIBUTION
Centralized text message reminders from a public insurance managed care organization have a small but signiﬁcant effect on increasing rates of the ﬁrst dose of human papillomavirus vaccine for parents with working phones capable of receiving text messages.
Registered with Clinicaltrials.gov (NCT01806714) on March 5, 2013. Sponsor: University of Rochester. * Address correspondence to: Cynthia M. Rand, M.D., M.P.H., Department of Pediatrics, University of Rochester School of Medicine and Dentistry, Box 777, 601 Elmwood Avenue, Rochester, NY 14642. E-mail address: [email protected]
C.M. Rand et al. / Journal of Adolescent Health 56 (2015) S17eS20
Human papillomavirus (HPV) vaccine has been recommended for adolescent females and males since 2007 and 2011, respectively [1,2]. However, only 57.3% of female adolescents aged 13e17 years and 34.6% of male adolescents aged 13e17 years had received at least one dose of the vaccine in 2013, and rates have plateaued in female adolescents . Practice-based remindererecall systems for vaccines due or missed have been shown to increase immunization rates [4e8] and have traditionally been done by mail or phone. Recently, text message reminders have been shown to improve childhood and adolescent immunization rates in a single urban setting [9e11]. However, only 16% of pediatric practices nationally use remindererecall systems . Barriers to use of such systems include cost, competing demands, concerns about lack of completeness of immunization and contact records, and insufﬁcient experience [13e16]. Centralized mail and phone remindererecall from a managed care organization (MCO) and a state immunization registry have been shown to increase rates of immunizations [17,18] and take advantage of economies of scale. Therefore, we conducted a randomized controlled trial (RCT), based in a large MCO, to evaluate the effectiveness of a centralized text messageebased remindererecall system on improving rates of the ﬁrst dose of HPV vaccination among low-income adolescents. We hypothesized that centralized textebased reminders would improve HPV immunization rates.
adolescent and excluded siblings. Patients were excluded if they transferred out of a participating practice or were no longer insured by the MCO during the study period. Intervention The MCO programmer reviewed vaccination data and sent text messages (up to four) to parents of eligible adolescents in the intervention group, using a third-party vendor who specializes in mobile communication. An initial notice informed parents that they were enrolled in a health message program from their child’s insurer: “MPHealth: Ur opted in 4 health messages frm ur child’s insurance [name of insurance plan] THX! Msg&Data Rates May Apply Text STOP to opt-out.” If participants did not opt out, a reminder for an HPV vaccination (MPHealth:‘Your [age] yr old is due for an HPV vaccine. Pls call [phone number and ofﬁce name] to schedule an appt. Text STOP to opt-out)’ was sent. Parents of participants who received HPV vaccine dose 1 during the study and so were eligible for doses 2 or 3 (based on recommended intervals) also received messages for those doses. If a patient was not yet eligible for the next dose based on billing and registry data, no message was sent. The MCO received a report from the thirdparty vendor identifying whether text messages were received. Controls
Methods Setting The study was based at the Monroe Plan for Medical Care, a large not-for-proﬁt MCO in upstate New York serving patients insured by Medicaid or the NY State Child Health Insurance Program. Members enroll in the MCO and are asked to update their contact information annually. Study design From July 2013 to March 2014, we conducted an RCT comparing text message reminders for HPV vaccination versus general adolescent health text messages, randomizing adolescents within each practice. Stata (Version 13.1; StataCorp, College Station, TX) was used to generate a randomization table. The study was approved by the University of Rochester’s Research Subjects Review Board. Participating practices We included 39 primary care practices (29 pediatric and 10 family medicine), each with more than 175 adolescents enrolled in the MCO from their practice. Practice managers received a letter on MCO letterhead notifying them of the intervention. Subjects The target population included adolescents aged 11e16 years enrolled in the Monroe Plan on July 1, 2013, with a primary care provider in a participating practice and with a phone number listed in the insurer’s database. Adolescents were eligible if they had no record of any HPV vaccinations in either billing data or the state’s immunization registry (NY law requires registry reporting of all vaccinations for patients aged less than 19 years). For families with multiple children, we randomly allocated a referent
Control group parents received the same initial message, followed by a control message about a different general adolescent health topic each time reminders were sent to the intervention group (e.g., ‘MPHealth: A healthy breakfast is associated with improved brain function, fewer missed school days, and improved mood for teens, Text STOP to opt-out’). Measures The main outcome was receipt of the ﬁrst dose of HPV vaccine, but secondary measures included subsequent HPV vaccine doses (i.e., receipt of HPV vaccine doses 2 and 3). Analyses We determined the rate of HPV vaccine receipt for intervention versus control group patients from the time of the ﬁrst message sent to the end of the study with the KaplaneMeier failure function. We determined hazard ratios using a clustered stratiﬁed Cox model with the Efron method to handle tied events and the Huber/White variance estimator that clustered on primary care provider and stratiﬁed on practice. We examined results for all subjects and separately for those with a valid phone number able to receive text messages, stratiﬁed by age group (11e13 and 14e16 years) and gender. Because our planned stratiﬁed analyses were limited by insufﬁcient sample size, we performed additional post hoc analyses that included all subjects, but controlled for age group and gender. Results We sent messages to the parents of 3,812 adolescents (descriptive statistics shown in Table 1). Slightly more than half were male and were insured by Medicaid; 74% were seen in pediatric practices. Almost half had a phone number that never received health or reminder text messages because of a phone that was not able to receive texts or not in service (760 of controls
C.M. Rand et al. / Journal of Adolescent Health 56 (2015) S17eS20 Table 1 Demographics of adolescent participants
N Age, years 11 12 13 14 15 16 Male Insurance Medicaid SCHIP Practice specialty Pediatric Family medicine Residence Urban/suburban Rural Messages received 0a 1 2 3 4
Control, n (%)
Intervention, n (%)
407 380 324 288 270 250 1,051
(21.2) (19.8) (16.9) (15.0) (14.1) (13.0) (54.8)
428 341 307 281 271 265 1,030
1,123 (58.5) 796 (41.5)
1,118 (59.1) 775 (40.9)
1,428 (74.4) 491 (25.6)
1,392 (73.5) 501 (26.5)
1,143 (72.6) 431 (27.4)
1,112 (71.2) 451 (28.9)
893 222 264 404 136
(46.5) (11.6) (13.8) (21.1) (7.1)
869 220 269 406 129
Table 2 Vaccination rate at the end of study period for subjects with valid phone numbers, by index patient sex and age groupa Vaccine
(22.6) (18.0) (16.2) (14.8) (14.3) (14.0) (54.4)
(45.9) (11.6) (14.2) (21.5) (6.8)
HPV ¼ human papillomavirus; SCHIP ¼ State Child Health Insurance Program. a Includes 133 control and 139 intervention participants who opted out after the introductory message.
and 730 of intervention) or opted out after the initial introductory message (133 of controls and 139 of intervention). During the study, 278 of controls and 205 of the group intervention opted out overall. Human papillomavirus vaccination rates When all subjects (with or without valid phone numbers) were included, there were no differences in HPV vaccination rates for any dose (15% of females and 14% of males in the control group; 14% and 16%, respectively, in the intervention group had 1 dose, data not shown). Similarly, when only subjects with a valid phone number capable of receiving text messages, who did not opt out after the introductory message were included in the stratiﬁed analysis (Table 2), there were no signiﬁcant differences for any HPV vaccine doses (1, 2, or 3) for either gender or age group. In a post hoc analysis for all participants (i.e., all ages and both genders) with valid phone numbers who did not opt out after the introductory message, controlling for age and gender (Table 3), there was a signiﬁcant increase of 30% in HPV vaccine dose 1 rates (13% of the control and 16% of the intervention group had received one dose). Discussion In this RCT of centralized MCO-generated text message reminders to the parents of adolescents due for the ﬁrst dose of HPV vaccine, we found a 30% increase in the ﬁrst-dose vaccination rates when the entire population with working phone numbers was included but no statistically signiﬁcant increases in HPV vaccination rates for age and gender groups. One major challenge was that only half of our sample had a working phone number able to receive text messages, so that our stratiﬁed analyses, which showed no signiﬁcant change in rates, had relatively small sample sizes.
Females aged 11 to <14 years HPV dose 1 498 11 HPV dose 2 533 4 HPV dose 3 504 1 Females aged 14 years HPV dose 1 386 9 HPV dose 2 394 2 HPV dose 3 380 1 Males aged 11 to <14 years HPV dose 1 601 15 HPV dose 2 636 7 HPV dose 3 600 2 Males aged >14 years HPV dose 1 440 11 HPV dose 2 458 4 HPV dose 3 440 1
13 3 1
1.3 (.9e1.9) 1.1 (.6e1.9) 1.9 (.8e5.0)
.20 .75 .17
11 7 1
1.1 (.5e2.4) 2.9 (.9e9.4) 2.8 (.6e13.7)
.72 .09 .19
20 8 2
1.3 (.9e2.0) 1.3 (.8e2.3) .6 (.2e2.5)
.18 .34 .51
13 3 1
1.3 (.8e2.1) 1.0 (.4e2.5) .9 (.2e4.8)
.32 .99 .89
HPV ¼ human papillomavirus. a KaplaneMeier failure function, hazard ratios from stratiﬁed Cox models. The models are stratiﬁed on practice and adjusted for standard errors by clustering on physician.
Studies show remindererecall is generally effective, yet few studies have examined HPV vaccine reminders speciﬁcally. We found a signiﬁcant improvement in the ﬁrst-dose HPV vaccination rates with remindererecall from the insurer for those with a valid phone number capable of receiving text messages. However, rates of one dose of HPV vaccine at the end of the study were 16%, compared to 13% in the control group, which is a modest improvement. Of note, these rates are lower than national averages because we selected only adolescents who had never received an HPV vaccine at the beginning of the study. A previous study of networked practices showed that parental telephone reminders were helpful for the second and third doses of HPV vaccine but minimally effective for dose 1 . Our ﬁndings, combined with those from other studies , suggest that ofﬁce-based interventions are also necessary to dramatically improve HPV vaccination rates , likely because of the issue of parental hesitancy to initiate HPV vaccination. Prior studies have emphasized the importance of a strong provider recommendation [20,21] so combining practice-based interventions with centralized reminders may be the most effective in increasing HPV vaccination rates. Because few practices implement remindererecall, centralized systems may be necessary to expand its use and reach Healthy People 2020 goals for adolescent immunization. Results from this and our prior study  show that MCO-based vaccine
Table 3 Vaccination rate at the end of study period for subjects with valid phone numbersa Vaccine
Male and female, aged 11e16 years
HPV1 HPV2 HPV3
1,925 2,021 1,924
13 6 2
16 7 2
1.3 (1.0e1.6) 1.2 (.9e1.7) 1.3 (.7e2.6)
.04 .27 .43
HPV ¼ human papillomavirus. a KaplaneMeier failure function, hazard ratios from stratiﬁed Cox models. The models are stratiﬁed on practice and adjusted for standard errors by clustering on physician and has an independent variable for age (treated as categorical) and gender.
C.M. Rand et al. / Journal of Adolescent Health 56 (2015) S17eS20
reminders can increase immunization rates but only by a limited amount. In addition, parents are generally receptive to centralized reminders . Although this study focused on the ﬁrstdose reminders, we postulate that once that dose of vaccine has been accepted, reminders for follow-up doses are more likely to be successful. Thus, future studies should evaluate methods to raise both HPV vaccine initiation and booster doses. Because immunization rates generally increase over time and most interventions on adolescent immunization rates have yielded only incremental beneﬁts , we see the need for further RCTs which are better than other study designs in identifying the true impact of new interventions. Our study highlights the role of secular trendsdboth intervention and control groups had a more than 10% increase in rates of the ﬁrst vaccine dose over the study period. Text messaging is a novel way for parents to receive health reminders and have been shown to be effective in improving rates of Tdap and meningococcal vaccine for adolescents  and in decreasing the time between HPV vaccine doses . In addition, many parents are interested in receiving text message reminders from their child’s doctor . However, health systems may be concerned about the need for parental consent to receive texts and transmission of electronic personal health information (ePHI). To avoid these concerns, we recommend obtaining consent for text message reminders at enrollment into the health system and not including ePHI in the message. Other options are to encrypt ePHI and require recipient authentication of the message. We expect that technology and security for texting in health care will continue to evolve as health systems realize its potential to improve communication. Limitations This study was conducted in a single MCO, which may limit its generalizability. Second, almost half of parents either did not have a phone capable of receiving texts or had a nonworking phone number. Although mail is more expensive, future studies could consider mixing an initial text message followed by a mailed reminder for those parents who are unable to receive text messages. In conclusion, we found a small but signiﬁcant difference in the ﬁrst-dose HPV vaccination rates for adolescents whose parents received text message reminders from their insurer for those parents with working cell phones capable of receiving text messages. Future studies should focus on raising the ﬁrst-dose HPV vaccination rates even further and on follow-up doses to achieve full protection against HPV-related disease. Funding Sources Funding for this project was provided by the Society for Adolescent Health and Medicine (SAHM) under a grant to SAHM from Merck. This work was presented in part at a SAHM grantee meeting in Cincinnati, Ohio on October 15, 2013.
References  Markowitz LE, Dunne EF, Saraiya M, et al. Quadrivalent human papillomavirus vaccine: Recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Recomm Rep 2007;56:1e24.  Centers for Disease Control and Prevention. Recommendations on the use of quadrivalent human papillomavirus vaccine in malesdAdvisory Committee on Immunization Practices (ACIP), 2011. MMWR Morb Mortal Wkly Rep 2011;60:1705e8.  Stokley S, Jeyarajah J, Yankey D, et al. Human papillomavirus vaccination coverage among adolescents, 2007e2013, and postlicensure vaccine safety monitoring, 2006e2014-United States. MMWR Morb Mortal Wkly Rep 2014;63:620e4.  Szilagyi PG, Schaffer S, Barth R, et al. Effect of telephone reminder/recall on adolescent immunization and preventive visits: Results from a randomized clinical trial. Arch Pediatr Adolesc Med 2006;160:157e63.  Szilagyi PG, Schaffer S, Shone L, et al. Reducing geographic, racial, and ethnic disparities in childhood immunization rates by using reminder/ recall interventions in urban primary care practices. Pediatrics 2002;110: e58.  Fiks AG, Grundmeier RW, Mayne S, et al. Effectiveness of decision support for families, clinicians, or both on HPV vaccine receipt. Pediatrics 2013;131: 1114e24.  Jacobson Vann JC, Szilagyi P. Patient reminder and patient recall systems to improve immunization rates. Cochrane Database Syst Rev 2005:CD003941.  Suh CA, Saville A, Daley MF, et al. Effectiveness and net cost of reminder/ recall for adolescent immunizations. Pediatrics 2012;129:e1437e1445.  Stockwell MS, Kharbanda EO, Martinez RA, et al. Text4Health: Impact of text message reminder-recalls for pediatric and adolescent immunizations. Am J Public Health 2012;102:e15e21.  Stockwell MS, Kharbanda EO, Martinez RA, et al. Effect of a text messaging intervention on inﬂuenza vaccination in an urban, low-income pediatric and adolescent population: A randomized controlled trial. JAMA 2012;307: 1702e8.  Kharbanda EO, Stockwell MS, Fox HW, et al. Text message reminders to promote human papillomavirus vaccination. Vaccine 2011;29:2537e41.  Tierney CD, Yusuf H, McMahon SR, et al. Adoption of reminder and recall messages for immunizations by pediatricians and public health clinics. Pediatrics 2003;112:1076e82.  Saville AW, Albright K, Nowels C, et al. Getting under the hood: Exploring issues that affect provider-based recall using an immunization information system. Acad Pediatr 2011;11:44e9.  Albright K, Saville A, Lockhart S, et al. Provider attitudes toward publicprivate collaboration to improve immunization reminder/recall: A mixed-methods study. Acad Pediatr 2014;14:62e70.  Pereira JA, Quach S, Heidebrecht CL, et al. Barriers to the use of reminder/ recall interventions for immunizations: A systematic review. BMC Med Inform Decis Mak 2012;12:145.  Hambidge SJ. The challenge of reaching adolescents to promote health care. Acad Pediatr 2013;13:184e5.  Szilagyi PG, Albertin C, Humiston SG, et al. A randomized trial of the effect of centralized reminder/recall on immunizations and preventive care visits for adolescents. Acad Pediatr 2013;13:204e13.  Kempe A, Saville A, Dickinson LM, et al. Population-based versus practicebased recall for childhood immunizations: A randomized controlled comparative effectiveness trial. Am J Public Health 2013;103:1116e23.  Patel A, Stern L, Unger Z, et al. Staying on track: A cluster randomized controlled trial of automated reminders aimed at increasing human papillomavirus vaccine completion. Vaccine 2014;32:2428e33.  Rosenthal SL, Weiss TW, Zimet GD, et al. Predictors of HPV vaccine uptake among women aged 19-26: Importance of a physician’s recommendation. Vaccine 2011;29:890e5.  Brewer NT, Fazekas KI. Predictors of HPV vaccine acceptability: A theoryinformed, systematic review. Prev Med 2007;45:107e14.  Saville AW, Beaty B, Dickinson LM, et al. Novel immunization reminder/ recall approaches: Rural and urban differences in parent perceptions. Acad Pediatr 2014;14:249e55.  Kharbanda EO, Stockwell MS, Fox HW, et al. Text4Health: A qualitative evaluation of parental readiness for text message immunization reminders. Am J Public Health 2009;99:2176e8.