An interactive voice response (IVR) system runsaudio scripts on a computer connected to a telephone.Because a script can be programmed tooutput diverse verbal instructions and accept input fromany key on a touch-tone telephone, this technology ishighly flexible. Although other technologies such as theInternet can perform similar functions with visual aswell as audio interactions, the telephone is more widelyavailable. Another advantage of IVR is that patients mayeither call or be called.
IVR appeals to healthcare organizations as a methodto promote health behavior and reduce costs by replacinghuman labor with computer technology. Someorganizations report saving hundreds of thousands ofdollars by using IVR.1 Before adopting this technology,however, organizations should consider several basicquestions: What is the best use for IVR? How do patientslike IVR? Is IVR cost-effective?
In healthcare, IVR has been used to promote a widevariety of health behaviors. These include remindercalls for medical appointments and testing, patient education,informing patients of test results, collectingpatient care data such as blood glucose levels, performingpatient surveys, and even providing care for specificdiseases such as obsessive compulsive disorder.2
The study by Polinski et al in this issue of the illustrates the use of IVR to remind patients about theneed for osteoporosis testing.3 Conducted at JeffersonMedical College, the Polinski study found that IVR didlittle to increase patient requests for testing. Although all ofthe possible reasons for IVR ineffectiveness were not measured,one major factor was difficulty contacting 50% of thepatients by telephone. By contrast, a review by Corkreyand Parkinson found that 6 of 7 studies showed a positiveeffect using IVR for appointment reminders and 3 of 4 studiesshowed a positive effect for immunization reminders.2Another study from Jefferson Medical College found thatIVR reminder calls increased rates of influenza immunizations,Pap tests, and mammograms.4 One lesson fromthese studies is that IVR effectiveness needs to be testedand monitored by an organization for each application.
IVR has both positive and negative benefits. For confidentialmaterial, people may be more willing to interactwith a computer than a person. In a study of counselingrelated to human immunodeficiency virus (HIV) testing,85% of patients chose an IVR session rather than talkingto a counselor on the telephone.2 In another study ofalcohol and drug consumption, IVR self-report rates wereconsistently higher than those achieved by a personaltelephone interview.5 For other types of data, such asgeriatric functional status, IVR identified fewer functionaldisabilities than did a personal visit.2
The response rates for IVR surveys are reported to belower than other survey techniques.5 This may be partlybecause individuals have fewer inhibitions abouthanging up on a computer. In a study comparing IVRwith mail and Internet for a physician satisfaction survey,the response rate for IVR was 34.7%, compared with50.8% by mail and 18.4% by Internet.6 Overall acceptabilityseems to be good, although less for IVR than forface-to-face contact in some studies.2
Because IVR is used for a wide variety of applications,blanket statements cannot be made about cost-effectiveness.At Methodist Medical Center of Illinois, we foundthat an IVR telephone appointment reminder systempaid for itself in less than 6 months by decreasing theno-show rate at a residency clinic from approximately25% to 17%. However, another study found mail morecost-effective than IVR to do a survey.6 As with overalleffectiveness, cost-effectiveness needs to be evaluatedfor each IVR application.
In summary, like other programmable technologies,IVR use is limited only by imagination. Most studies havefound improved health behaviors by using IVR. Carefulattention to design and testing of systems is important tomake IVR a cost-effective solution.
From the University of Illinois College of Medicine at Peoria, Ill.
Address correspondence to: David E. Trachtenbarg, MD, Medical Director, InformationTechnology, Methodist Medical Center of Illinois, 221 NE Glen Oak Ave, Peoria, IL 61636.E-mail: email@example.com.
Health Manag Technol.
1. Bukowski B. What works: dietary IVR (interactive voice recognition) nets a millionin savings for NY hospital. 1998;19:30.
Behav Res Methods Instrum Comput.
2. Corkrey R, Parkinson L. Interactive voice response: review of studies 1989-2000.2002;34:342-353.
Am J Manag Care.
3. Polinski. JM. Interactive voice reponse telephone calls to enhance bone mineraldensity testing. 2006;12:321-325.
Am J Med Qual.
4. Crawford AG, Sikirica V, Goldfarb N, et al. Interactive voice response remindereffects on preventive service utilization. 2005;20:329-336.
Behav Res Methods
5. Corkrey R, Parkinson L. A comparison of four computer-based telephone interviewingmethods: getting answers to sensitive questions. 2002;34:354-363.
6. Rodriguez HP, von Glahn T, Rogers WH, Chang H, Fanjiang G, Safran DG.Evaluating patients' experiences with individual physicians. A randomized trial ofmail, Internet, and interactive voice response telephone administration surveys.2006;44:167-174.