Sexually transmitted diseases (STDs) are a significant and increasing health risk in the United States. The Centers for Disease Control and Prevention estimates that 19 million new STDs occur every year in this country.1 From 2007 to 2011, reported STD cases in Florida increased by 18%, exceeding 99,000 cases in 2011, and setting a record for the highest STD total ever recorded in the state. Prompt notification of positive test results and treatment could slow the spread of STDs.
Florida STD staff in county health departments (CHDs) primarily use “traditional methods” of telephone calls and home visits to encourage clients to return to the clinic for treatment. If staff are not able to reach positive clients in a timely manner, there may be delays in notification and treatment and consequently possible ongoing spread of STDs. Significant staff time is spent in locating positive clients for follow-up; therefore, presumptively treated positive clients and clients who test negative are not contacted by field staff.
Innovative methods are needed to intervene in the spread of disease and to use staff time effectively.2 Texting is a potentially viable method of communicating health information because of its wide use for general communications.3,4 It also has the potential to address racial/ethnic disparities, as mobile phone ownership and the use of texting has been found to be very high among racial/ethnic minorities.3,5–7 Texting has been used to send health education messages to clients,8–10 prompt clients to take their medication,11–14 and remind clients of appointment dates and times.15 Texting has increased client-provider communication in chronic disease management16 and improved operational efficiencies in STD notification efforts in UK clinics.17–19
Some US STD programs have drafted policies and protocols for the use of texting to facilitate client-provider communication in STD intervention,20–22 but no state STD programs have. In 2009, Peoria CHD in Illinois began delivering gonorrhea and chlamydia results using coded texts. They reported a decrease in the time required for follow-up with clients and lowered program costs, as a result of less time spent by staff locating clients.23 Starting in December 2010, Clay CHD in Florida implemented Peoria CHD’s coded texting project and also documented similar outcomes to Peoria.24 Neither the Peoria nor Clay CHD texting projects have been rigorously evaluated.
Implementation of the Florida STD Texting Project
The Florida Department of Health STD Program adapted the texting protocols of Peoria and Clay CHDs to be used throughout the state. Peoria and Clay CHDs manually entered texts into Outlook, whereas Florida’s electronic STD surveillance application, the Patient Reporting Investigation Surveillance Manager (PRISM),25 was modified to allow standardized texts to be sent to clients. The Florida Department of Health STD Program assisted Florida CHDs in developing their policies and procedures, and trained them on how to obtain consent and to send texts. Individual CHDs determined the content of the texts and instruction cards, obtained consent, sent texts, and received call backs. County health departments could offer texting of chlamydia (positive and negative), gonorrhea (positive and negative), and syphilis (negative only) results.
Within the CHDs, the texting process begins when clients are offered the option to receive test results through text or traditional methods at client intake. If they agree to texting, they add their cell phone number and carrier to the consent form for medical treatment. Clients are given an instruction card that explains the meaning of the coded texts (see Fig. 1 for the content of texts). Information from the consent form is entered into the client’s PRISM record. Once laboratory results enter PRISM, a button is clicked and the client receives a text within seconds. Only one code is sent to clients to minimize confusion. Clients negative for chlamydia, gonorrhea, and syphilis are texted a code 2. Clients positive for either chlamydia or gonorrhea are texted a code 1 and given a phone number to call for treatment referral. If the client is positive, 2 reminder texts are sent automatically by PRISM, 48 and 72 hours after the first text. Because of the added complexity of determining a syphilis positive result (confirmatory testing and staging of the disease), clients positive for syphilis are notified through traditional methods. HIV test results are given to clients using traditional methods due to state laws requiring in-person notification for HIV.
Between November 2011 and January 2012, Clay CHD piloted the texting features in PRISM. After January, texting was available to all Florida CHDs. Texting was not available to private clinics. Within the first year of the texting project, Clay, Duval, Seminole, Escambia, Miami-Dade, and Orange County began texting out of all 67 Florida counties.
The aims of this descriptive evaluation were to assess (1) client uptake and completion of texting and (2) whether texting was associated with a shorter treatment time frame.
MATERIALS AND METHODS
The evaluation period was from February 2012 through January 2013. Data from Clay, Duval, and Seminole CHDs were evaluated, as these were the only CHDs with a full year of data collection. Outcome results reflect a cross-sectional sample of clients who either self-selected into the texting (texters) or the traditional notification (nontexters) groups. Only Seminole CHD offered texting of syphilis results throughout the evaluation period; therefore, only chlamydia and gonorrhea results from all 3 counties were included for analysis. This evaluation did not qualify as research and did not require an institutional review board approval. Descriptive statistics were computed using SPSS (PASW Statistics 18). Client characteristics analyzed included the county where the client was tested, sex, age, race, morbidity status, and if they were presumptively treated.
To assess client uptake and completion of texting, we looked at several process outcomes: (1) acceptability of the texting notification option in the form of opt in, (2) client responsiveness to text notification in the form of call backs, (3) text transmission failure rate in the form of undeliverable texts, and (4) client understanding of texting instructions in the form of texted replies asking a question. Opt-in was calculated using the number of clients that opted into texting divided by the number of clients tested.
Clients were instructed to call the number in the text if they received a code 1 and not to call or return to the clinic if they received a code 2. To assess client responsiveness to text notification, the number of clients who called back by morbidity status was calculated. We also measured the time from when the client was first texted to when they called the health department back. We excluded clients who were presumptively treated because they may not have been as motivated to call the health department. Clients were considered presumptively treated if their treatment date was before the date the test result was received by PRISM (lab add date).
The text transmission failure rate was measured by the number of clients whose first text was returned as undeliverable divided by total clients texted. Client understanding of texting instructions was assessed by looking at the number of clients that texted back a question divided by total clients texted.
To assess whether texting decreased treatment time frames, we calculated the time from the lab add date to client treatment date and compared this by notification status (texter vs. nontexter). We excluded presumptively treated clients. To evaluate if calling back the health department was associated with treatment time frames, we also compared the time from lab add date to treatment date for texted clients by call back status. Mann-Whitney U tests were conducted to compare means, and χ2 tests were conducted to compare proportions, with P values of less than 0.05 considered significant differences.
Clay, Seminole, and Duval CHDs offered texting to all clients visiting the clinic for chlamydia and/or gonorrhea testing, which from February 1, 2012, to January 31, 2013, was 10,272 clients (see Table 1). Overall, most clients were female (52%), older than 24 years (56%), black (54%), and negative for chlamydia and gonorrhea (83%). Characteristics varied greatly between the 3 CHDs (data not shown). Significant differences between nontexters and texters were found for all demographic variables, except for black clients (all P < 0.016).
Acceptability of Texting Notification
The proportion of clients that opted in was 52% (county range was 40%–73%). Opt-in was highest among clients who were positive for chlamydia and/or gonorrhea (58%), female (58%), younger than 25 years (55%), and white (54%; see Table 1). The only groups with opt-in levels below 50% were men and those whose race was other or unknown.
Client Responsiveness Among Texters
Among nonpresumptively treated positive texters, 198 (57%, county range was 54%–59%) called back, and among negative texters, 240 (6%) called back (see Table 2). On average, positive texters called back in approximately 1 day, and negative clients called back in approximately 5½ days.
Transmission Failure Rate and Client Understanding of Texting Instructions
Two hundred eighteen (4%) texts failed to transmit to a client and were returned as undeliverable. Overall, 130 (2%) clients responded to the text we sent them with a texted question (90% were negative and 10% were positive). Client understanding was lowest for Duval CHD, with 106 (4%) clients texting back a question.
Treatment Time Frames
Overall, 553 (5%) clients were positive for gonorrhea and/or chlamydia and nonpresumptively treated at the time of diagnosis (see Table 3). Of these, 345 were texters and 208 were nontexters. Overall, texters received treatment in an average of 5.1 days compared with 6.7 days for nontexters (P = 0.036) and all positive clients, irrespective of texter status, were treated. The biggest difference was for Seminole (4.6 vs. 6.3, P = 0.024). Clay CHD had shorter treatment time frames among nontexters. Texters who called back were treated 3 days earlier than those who did not call back (P < 0.001). Significantly more texters were treated within 10 days than nontexters (88% vs. 80%, P = 0.015), and significantly more texters who called back the CHD were treated within 10 days than texters who did not call back (P = 0.001)
The Florida STD Texting Project is the first state-initiated project to directly deliver STD test results to clinic clients in the United States via text. It is also novel in that texting was integrated into the state’s electronic STD surveillance database. Our long-term goal is to offer texting of STD test results statewide to improve the timeliness of STD diagnosis, notification, and treatment and to reduce staff burden. Clay, Seminole, and Duval CHDs were the first to start texting, and they represent small-, medium-, and large-sized counties, respectively, in Florida. Among texters, there were significantly more clients who were female, younger than 25 years, white, positive for gonorrhea and/or chlamydia, and presumptively treated than among nontexters.
Overall, opt-in was 52%. The higher opt-in among clients younger than 25 years and who are female may reflect a broader trend of frequent texting among these groups.3,26 There are a number of reasons why texting may have been appealing to clients. Unlike with traditional notification, texting allowed negative clients to obtain their results without returning to the clinic. We can presume that this saves time and transportation costs, although this was not measured. Positive clients who were presumptively treated also received their results, whereas with traditional notification, they do not.
Texters who called back received treatment significantly sooner than texters who did not call back, although we do not know if this is because texting decreased time to treatment or because texters were more motivated to seek treatment sooner. Overall, 57% of positive texters called back, and on average, they called back in 1 day. The response rate was 56% among STD clients of the Multnomah CHD, who were prompted with a text to call the health department to discuss results.27 By encouraging all positive texters to call back, we were able to encourage those not appropriately presumptively treated to return to the clinic when they may have not otherwise. Most negative clients did not call back, but a small proportion (6%) did. They may represent the “worried well” or clients who had forgotten the texting instructions. Our proportion of undeliverable texts was nearly equal to that of a project in the United Kingdom, where clients were texted their STD results.17
Overall, there was approximately a 1½ day decrease in treatment time frames for texters (P = 0.036). It is unclear if a day and a half difference has clinical significance, but it is meaningful in terms of reducing the amount of time that an infected person is potentially out in the community infecting others. In addition, one of Florida’s objectives for its State Health Improvement Plan is to treat 90% of women infected with a bacterial STD within 14 days of specimen collection by 2016.
An important lesson learned was that treatment time frames were not a straightforward measure of the success of texting. The CHD notification strategies, caseload, and available staff resources may have impacted treatment time frames. Clay CHD’s notification strategy was to give texters a 2-day grace period to return to the clinic on their own before calling them, whereas they called nontexters immediately. These different notification strategies and their low testing volume may have been the reason that their nontexters were treated sooner. Despite this outcome, Clay CHD staff felt that waiting 2 days saved valuable staff time. Duval CHD had the same notification strategy as Clay CHD, but Duval CHD texters received treatment sooner. A high caseload in Duval CHD may have prevented staff from calling nontexters immediately.
In contrast to Clay and Duval CHDs, Seminole CHD’s notification strategy was to call both positive texters and nontexters immediately. This did not save staff effort, but texting may have augmented traditional notification methods, as Seminole CHD saw the greatest difference in treatment time frames. The lack of difference in treatment time frames for those that called back and those that did not in Seminole may also be explained by the fact that they called everyone as soon as possible, which gave clients less of a chance to call in themselves. Programs that need to conserve and focus staff effort for “difficult” to contact clients may wish to give texts a few days to work (as in Duval and Clay CHDs). Programs with the necessary available staff and who wish to prioritize the reduction of treatment time frames may elect to add texting to traditional notification methods (as in Seminole CHD).
There were several other local barriers we found anecdotally that may have impacted the implementation of texting, including: texting not being a priority, programs not wanting to be the “first” to try texting, staff fears of increased workload, delays in obtaining approval from local leadership, and difficulties in determining staff responsibilities for the texting process. In addition, we received some anecdotal reports of phone carriers delivering incomplete messages.
This texting project is the first to use an in-house surveillance application, PRISM, to transmit texts. The use of PRISM has limitations and advantages. Limitations include the following: (1) texting through PRISM requires specific user privileges; (2) PRISM is not a clinic management application; (3) PRISM is not currently used in every CHD; and (4) PRISM produces quantitative outcomes and cannot explain qualitative elements. The advantages include the following: (1) PRISM tracks texts and call backs with minimal staff effort; (2) PRISM measures the same outcomes across counties; (3) it lessens the burden on local programs to collect their own outcome data; (4) texting data for each client forms a part of their PRISM case management record; and (5) given the distinct operations of each local STD program, PRISM was invaluable in standardizing procedures and keeping everyone accountable. In addition, the implementation of texting through PRISM can be expanded to 13 other states. PRISM is currently used in 8 states, and 5 other state health departments are in the process of transitioning to it.
There are limitations to this evaluation. Our sample was a convenience sample and may not be generalizable to other populations or geographic areas. The 3 CHDs included in the evaluation may not be representative of all 67 counties in Florida, but they have the advantage of representing both urban and suburban client populations. The data are cross sectional so causality cannot be attributed. It is possible that those who self-selected into the texter group were more motivated to obtain treatment. The frequent use of presumptive treatment by CHDs also reduced our ability to measure changes in treatment time frames. We could not assess the quality of education around texting that clients received at each CHD, how many clients did not have cell phones or could not text, and reasons clients opted out. We excluded from analysis clients who were only texted syphilis results, and there may be significant differences in that group in comparison with those included in this evaluation. We were not able to collect data only available at the local level, such as clinic volume, but anecdotally, Seminole experienced a 28% decrease in negative clients returning compared with the prior year. It is possible that the other CHDs saw similar decreases in clinic burden to Seminole. Although not feasible for this evaluation, future research should assess how much staff and client time is saved using texting.
This evaluation shows the feasibility of using texting for STD results notification across distinct programs. Ultimately, STD programs should use the communication tools that clients prefer. Nationally, 98% of texts are read and they are typically read within 15 minutes of being received.28 Given the hours, days, and even weeks that STD programs invest in tracking down positive clients, texting also has the potential to save valuable health department resources. When carried out as intended (i.e., clients opted in, called back, and returned for treatment) clients obtained treatment approximately 3 days sooner than clients who did not opt in. This reduced the time staff spent contacting clients and potentially reduced the further spread of STDs within the community. However, texting is not a cure-all. It still requires the participation of health department staff and clients to be actively engaged in the process.
1. Centers for Disease Control and Prevention. Sexually Transmitted Disease Surveillance 2011. Atlanta, GA: Centers for Disease Control and Prevention, 2012.
2. Golden MR, Kerndt PR. Improving clinical operations: Can we and should we save our STD clinics? Sex Transm Dis 2010; 37: 264–265.
4. Blumberg SJ, Luke JV. Wireless Substitution: Early Release of Estimates From the National Health Interview Survey, January–June 2011 [National Health Interview Survey Web site]. (June). Available at: http://www.cdc.gov/nchs/nhis.htm
. Accessed May 15, 2013.
7. Samal L, Hutton HE, Erbelding EJ, et al. Digital divide: Variation in internet and cellular phone use among women attending an urban sexually transmitted infections clinic. J Urban Health 2009; 87: 122–128.
8. Lim MSC, Hocking JS, Aitken CK, et al. Impact of text and email messaging on the sexual health of young people: A randomised controlled trial. J Epidemiol Community Health 2012; 66: 69–74.
9. Mbuagbaw L, Thabane L, Ongolo-Zogo P, et al. The Cameroon mobile phone SMS (CAMPS) trial: A protocol for a randomized controlled trial of mobile phone text messaging versus usual care for improving adherence to highly active anti-retroviral therapy. Trials 2011; 12: 1–8.
10. Levine D, McCright J, Dobkin L, et al. SEXINFO: A sexual health text messaging service for San Francisco youth. Am J Public Health 2008; 98: 393–395.
11. Hardy H, Kumar V, Doros G, et al. Randomized controlled trial of a personalized cellular phone reminder system to enhance adherence to antiretroviral therapy. AIDS Patient Care STDS 2011; 25: 153–161.
12. Harris LT, Lehavot K, Huh D, et al. Two-way text messaging for health behavior change among human immunodeficiency virus–positive individuals. Telemed J E Health 2010; 16: 1024–1029.
13. Lester RT, Ritvo P, Mills EJ, et al. Effects of a mobile phone short message service on antiretroviral treatment adherence in Kenya (WelTel Kenya1): A randomised trial. Lancet 2010; 376: 1838–1845.
14. Pop-Eleches C, Thirumurthy H, Habyarimana JP, et al. Mobile phone technologies improve adherence to antiretroviral treatment in a resource-limited setting: A randomized controlled trial of text message reminders. AIDS 2011; 25: 825–834.
15. Downer SR, Meara JG, Da Costa AC, et al. SMS text messaging improves outpatient attendance. Aust Health Rev 2006; 30: 389–396.
16. Fischer HH, Moore SL, Ginosar D, et al. Care by cell phone: Text messaging for chronic disease management. Am J Manag Care 2012; 18: e42–e47.
17. Dhar J, Leggat C, Bonas S. Texting—A revolution in sexual health communication. Int J STD AIDS 2006; 17: 375–377.
18. Menon-Johansson A, McNaught F, Mandalia S, et al. Texting decreases the time to treatment for genital. Sex Transm Infect 2006; 82: 49–51.
19. Kachur R, Adelson S, Firenze K, et al. Reaching patients and their partners through mobile: Text messaging for case management and partner notification. Sex Transm Dis 2011; 38: 149.
23. Mcclallen RB. Coded Text Messaging to STD Clients: A Step in the Right Direction. Poster Presented at: 2010 National STD Prevention Conference [P54]. Atlanta, GA.
25. Shiver SA, Schmitt K, Cooksey A. Use of a business approach to improve disease surveillance data management systems and information technology process in Florida’s Bureau of STD Prevention and Control. Public Health Rep 2009; 124: 98–102.
27. Mendez J, Maher J. Evidence supporting the use of text messaging for partner services. Sex Transm Dis 2012; 39: 238–9.