Texting Giardia Patients a Survey Link Saved Colorado 40 Staff Hours

Interviewing people who have certain infections — about their symptoms, jobs and recent exposures — is central to spotting and stopping outbreaks, and the sooner the better, before memories fade. But rising case counts (partly from newer tests such as gastrointestinal panels), competing emergencies and shrinking budgets have made it hard to keep up.

Expecting post–COVID-19 funding cuts and competing demands such as mpox and highly pathogenic avian influenza, the Colorado Department of Public Health and Environment (CDPHE) tested an interview method with no phone calls at all for people with Giardia infection.

Background

• Reporting Giardia infections is required in Colorado, but interviewing patients became optional in 2018, and many local health departments stopped. During 2016–2017, 75% of people with Giardia were interviewed by phone; in 2019, only 8%.
• Public health in Colorado is decentralized: during the study, 57 health jurisdictions served 64 counties. CDPHE ran the pilot for the 36 jurisdictions it was responsible for.
• CDPHE and the Colorado School of Public Health had created an Enteric Disease Interview Team (EDIT) to interview patients for local departments that ask for help. CDPHE now offers Giardia interviews only through the online survey.

How it worked

1. From September 1, 2023, to May 30, 2024, everyone with a lab-confirmed Giardia infection assigned to CDPHE got a personalized text in English and Spanish inviting them to reply "yes" to take a survey. Phone numbers came with the lab report or, if missing, from medical and vaccination records. Non-responders got one follow-up text after 24 hours.
2. Those who opted in received a unique survey link, sent through Twilio from inside REDCap, a HIPAA-compliant survey platform. Unfinished surveys triggered automatic reminders every 2 days, up to 3 times.
3. The survey had 32 questions for everyone — symptoms, contacts, travel, exposures, occupation, demographics and, for adults, sexual history — branching to as many as 203 depending on answers. Completing at least the core 23 counted as finishing.

Of 564 cases reported during the period, 341 were assigned to CDPHE. 20 were excluded (3 without lab results, 1 living outside Colorado, 16 with a co-infection), as were 16 people with no mobile phone. Texts went to 305 of 321 (95%) eligible people.

Results

From text invitation to completed survey. CDC.

• The median number of texts needed was one (range 1–4).
• 69% completed the survey within 1 day of the first text, and 99% within 6 days — well within CDPHE's 7-day deadline for Giardia interviews.
• Once they opened it, people took a median of 13 minutes; 95% finished within an hour.

Who completed it

Of those who finished, 65 (68%) also took a satisfaction survey: 89% used a mobile phone, 97% had no technical problems, and none wanted a language other than English or Spanish.

Data quality

93%–100% answered each of 11 key nonsensitive questions (for example, symptoms, ill household members, and work in health care, child care or food handling), and 90%–96% answered the sexual history questions.

Time saved

That's roughly 25 minutes saved per case and 40 hours overall — about one sixth of the time phone interviews would have taken.

What it means

• Text-plus-survey let CDPHE gather complete data on about a third of patients quickly and cheaply — people who otherwise would likely not have been interviewed at all. Its 31% completion is below the 75% phone interview rate of 2016–2017, and some groups were underrepresented, so raising opt-in deserves work.
• Online surveys may also be less affected by social desirability bias — answering what an interviewer seems to want — and may help collect sensitive information.
• The approach could be combined with phone calls for people who don't opt in. It may not suit high-consequence diseases that require immediate reporting, but could work for lower-consequence infections such as Giardia, Campylobacter, pertussis and West Nile virus; for pathogens with historically low interview response, such as Shigella and mpox; and for contact tracing, vaccination information or symptom monitoring during outbreaks such as avian influenza, Ebola or Marburg.

Limitations

1. The sample was small and limited to Giardia patients.
2. Non-responders weren't interviewed, so differences between them and responders couldn't be assessed.
3. The survey was relatively short, which may have raised completion compared with longer ones.

Sources

Based on Hewitson I, Tran A, Younger A, Jervis RH, "Evaluation of a Combined Text Messaging and Online Survey Protocol for Giardiasis Case Investigation — Colorado, September 2023–May 2024," Morbidity and Mortality Weekly Report, Centers for Disease Control and Prevention; a work of the United States government in the public domain. The imported copy was cut off in three places; the missing text was taken from the same report's full text in PubMed Central (PMC12270486).