Use and Impact of an Automated Telephone Outreach System for Asthma in a Managed Care Setting

Objective: To test the ability of an automated telephone outreach intervention to reduce acute healthcare utilization and improve quality of life among adult asthma patients in a large managed care organization.

Study Design: Randomized clinical trial.

Methods: Patients with persistent asthma were randomly assigned to telephone outreach (automated = 3389, live caller = 192) or usual care (n = 3367). Intervention participants received 3 outreach calls over a 10-month period. The intervention provided brief, supportive information and flagged individuals with poor asthma control for follow-up by a provider. A survey was mailed to 792 intervention participants and 236 providers after the intervention. Additional feedback was obtained as part of the final intervention contact.

Results: The intent-to-treat analysis found no significant differences between the intervention and usual-care groups for medication use, healthcare utilization, asthma control, or quality of life. Post hoc analyses found that, compared with the control group, individuals who actually participated in the intervention were significantly more likely to use inhaled steroids and to have had a routine medical visit for asthma during the follow-up period and less likely to use short-acting b-agonists. They also reported higher satisfaction with their asthma care and better asthma-specific quality of life. Of surveyed providers, 59% stated the program helped them to clinically manage their asthma patients and 70% thought the program should be continued.

Conclusions: This study did not find improved health outcomes in the primary analyses. The intervention was well accepted by providers, however, and the individuals who participated in the calls appeared to have benefited from them. These findings suggest that further studies of automated telephone outreach interventions seem warranted.

(Am J Manag Care. 2006;12:725-733)

Asthma is a common disease that places a substantial burden on affected individuals and the healthcare system.^1-3 The past decade has seen an increasing emphasis on population-based disease management programs for asthma.^4-6 Central to the efficient operation of such disease management programs is the ability to identify individuals at greatest risk for acute exacerbations. Large administrative and clinical databases offer 1 solution for this need.^7-12 Recently Vollmer et al¹³ and Peters et al¹⁴ showed that a simple index of asthma control¹⁵ was also predictive of future acute-asthma healthcare utilization and that this index had predictive value above and beyond what can be achieved by use of administrative data alone. The combination of information about prior healthcare utilization and current level of control thus has the potential to enhance disease management efforts.

Unfortunately, information about asthma control derives from participant self-reporting and so is not typically available from administrative healthcare databases. To guide healthcare interventions, automated telephone systems are an option for cost-effective gathering of such information. Such systems have been tested in multiple studies to collect health data, monitor patients, provide health education, and remind patients about appointments or health screening activities.¹⁶ These systems have the potential to reach large numbers of patients at relatively low cost,^17,18 and thus are a promising strategy to enhance chronic disease management.^18,19

This article describes the implementation of, and reactions to, a Centers for Disease Control and Prevention-funded study designed to test the ability of an automated telephone outreach intervention to reduce acute healthcare utilization and improve quality of life among adults with asthma in a large managed care organization. The results have implications for the use of such telephone outreach systems for a variety of disease management programs, not limited to asthma.

METHODS

Research Setting and Target Population

Kaiser Permanente Northwest (KPNW) is a large, group-model health maintenance organization centered in Portland, Oregon. Kaiser Permanente Northwest provides comprehensive, prepaid healthcare service, including access to inpatient, outpatient, and emergency services, to its approximately 65 000 members, whose demographic and socioeconomic characteristics correspond closely to those of the area population as a whole.²⁰ Kaiser Permanente Northwest uses a comprehensive electronic medical record to identify health plan members eligible for this study.

Individuals were eligible for the study if they were aged 18 years or older as of January 1, 2003, and were either (1) on the KPNW high-risk asthma registry or (2) had at least 180 days of antiasthma medication dispensings during the 2-year period 2000 through 2001 and at least 1 medical contact for asthma during the same 2 years. The high-risk asthma registry includes individuals who received 4 or more canisters of a short-acting b-agonist in the past 12 months and who did not have chronic obstructive pulmonary disease (COPD). Based on feedback from clinicians during the initial field-testing of the intervention, we also excluded 850 individuals who had COPD listed on their problem list (nominally a list of all chronic conditions a member is known to have). The resulting sample size was 6948.

Study Design

Eligible individuals were randomly assigned to either usual care (n = 3367) or telephone outreach (n = 3581). To test the reactions of members to receiving an automated phone call, 192 members of the telephone outreach group were randomly selected to receive the calls from a real person; the remaining 3389 received automated calls.

Intervention

The intervention was designed to provide brief, supportive information to all participants and to flag individuals with poor asthma control for further follow-up by the healthcare system. The intervention consisted of 3 rounds of calling about 5 months apart. The calls consisted of a series of questions (Appendix, available at www.ajmc.com.) designed to assess (1) recent emergency department or hospital care for which the member had not had a follow-up visit, (2) current level of asthma control,¹⁵ (3) current patterns of asthma medication use, and (4) whether the member could identify a primary care provider whom he or she usually saw for asthma care. We also asked if the patient had ever been told he or she had "emphysema or COPD" as a means of identifying COPD patients who were not flagged as such on the problem list. These patients received an abbreviated intervention but were kept on the call list unless a diagnosis of COPD was subsequently added to their problem list. Based on the responses to these initial questions, members were offered (optional) tailored feedback regarding their overall level of asthma control and their use of asthma medications. Feedback was designed to convey a positive message without being prescriptive. For example, an individual who reported waking up at night due to asthma symptoms would be told that this sleep disruption did not need to be the case–his or her asthma could be in better control–followed by a suggestion to discuss this situation at his or her next doctor visit. Similarly, a patient who reported not having an inhaled corticosteroid (ICS) prescription would hear about the importance of these medications for maintaining good asthma control. A person who reported using ICS for symptom relief only would hear that these medications work best when taken every day. Calls concluded with an offer to receive information about additional Kaiser Permanente resources followed by an offer to call again in 5 months.

The calls were designed to be brief and typically lasted less than 10 minutes. The calls were made using speech recognition (or speech-enabled) technology, which does not require the respondent to push buttons to respond. We hoped this more "normal" interaction would increase participation with the intervention. The automated and live-person calls used identical scripts; responses to the questions were computerized and used to generate 1 of a number of different text messages that were entered into the electronic medical record as telephone encounters. Participant responses to intervention questions were used to flag participants at high risk of a future exacerbation (Table 1), and an electronic alert was placed in the medical record of each flagged participant. These alerts required that the participant's provider review the encounter and clear the alert from the record. The hope was that this review would trigger some form of provider follow-up contact with the patient.

Finally, the intervention program identified patients who were not regularly seeing a consistent provider for their asthma care and assigned them a primary care provider. To mimic the results we might see in a real-world implementation, the intervention was designed to appear to members as a test program being evaluated by the health plan rather than as a research study. To this end, study investigators enlisted key members of the health plan to help design and promote the intervention.

Participants

The study was approved by the KPNW institutional review board. Informed consent was obtained by sending all intervention participants a letter at the outset of the study describing the intervention and giving them a toll-free number to call to opt out of the intervention. During the initial phone contact, the intervention was described again, and participants were given another chance to opt out. Individuals were dropped from the call list if they had previously refused to participate, had COPD added to their problem list, or had left the health plan. Additionally, the final call round was limited to individuals with at least 1 prior successful intervention contact. Finally, during the initial call, members were asked to confirm that they had asthma, and persons who denied having the condition were excluded from further calling.

Data Collection

Primary outcomes of the study related to healthcare utilization, medication use, and quality of life. Important secondary outcomes were the extent to which the intervention could be successfully implemented and the reaction to the intervention by patients and providers. The secondary outcomes were important because, regardless of how the intervention modality worked for asthma, if feasible it could be adapted for use for a variety of chronic conditions. We therefore collected extensive "process" data, including responses to a survey of providers whose patients were enrolled in the intervention; responses to surveys mailed to health plan members; feedback from patients obtained immediately after the third intervention call; and results of an audit of charts that had a flagged encounter to determine what, if any, follow-up occurred. We also tracked participation at each round of calling.

Information on baseline characteristics of the study participants came from 3 sources: administrative data on age and sex, the healthcare utilization data used to identify study participants, and a survey mailed to a random sample of 549 health plan members in November 2002, about 7 months before implementation of the intervention and prior to randomization. The response rate to the baseline survey was 83% and did not differ for eventual intervention and control participants. Follow-up data were similarly obtained from a survey of 1583 randomly selected participants 1 month after the final calling period. Response to the follow-up survey was 65% for participants in both treatment groups. We also collected healthcare utilization data on all participants from the start of the initial calling until 1 month after the last round of calling. The latter were reported only for individuals (90% of randomized) with at least 6 months of health plan eligibility after randomization. Survey outcomes included demographic data, measures of health status, current asthma control, patterns of medication use, asthma quality of life, self-management practices, attitudes about asthma, and satisfaction with care. Asthma control was measured using the Asthma Therapy Assessment Questionnaire (ATAQ) control index, a 5-point scale reflecting the number of control problems reported during the last month.¹⁵ Asthma quality of life was measured using the mini-Juniper Asthma Quality of Life Questionnaire,²¹ which gives rise to 4 subscales and an overall summary measure, and the Asthma Impact Survey, which was used with permission from QualityMetric Incorporated (Lincoln, RI). Abstracted healthcare utilization data included dispensings of various classes of antiasthma medications and occurrences of both routine and acute healthcare utilization.

Statistical Methods

The primary outcome analysis used an intent-totreat design that included in the intervention group all randomized individuals, as well as persons who declined to participate in the intervention. The study is thus best viewed as an effectiveness trial, because it attempts to measure the effect of using the intervention in the real world. Secondary analyses were conducted for individuals younger than 65 years (for whom the diagnosis of asthma may be more robust and the ATAQ instrument more predictive) and for individuals with more severe disease (defined as 9 or more asthma medication dispensing events, or the combination of 6 or more asthma medication dispensing events and at least 1 emergency department visit or hospitalization for asthma during 2002). We also present a post hoc analysis limited to individuals who actually participated in at least 1 intervention call. Per protocol, the 2 intervention arms (automated and live-person calling) are combined for the primary and post hoc analyses. Categorical data from the follow-up survey were compared using the Pearson ?² test, whereas continuous or pseudocontinuous data were compared using t tests. Similar procedures were used to compare, during the follow-up period, the proportion of individuals who were dispensed asthma controller medications and the number of canisters of various antiasthma medications that were dispensed. Finally, we also used Pearson χ² tests to compare the number of routine and acute (after hours, urgent care, emergency department, and hospitalization) visits for asthma during the followup period. For the post hoc analyses, we adjusted for potential confounders by including them as main effect covariates in multivariate linear and logistic regression models.

The analysis of process data relied mainly on the use of simple descriptive statistics, as well as Pearson χ² tests for comparing intervention participants assigned to automated versus live-person calling. All analyses were conducted using SAS version 8.2 (SAS Institute, Inc, Cary, NC). The term "significant" refers to a P value of less than or equal to 0.05, and all P values were 2-sided.

RESULTS

Baseline Demographics

The usual-care and intervention groups were comparable at baseline (Table 2). The mean age of the participants was 52 years (range, 18-99 years). Sixty-five percent were women. Consistent with the demographics of the population served by KPNW, the vast majority of participants (mean, 92%) were white and not of Hispanic origin. About 50% had never smoked. Use of emergency health services for asthma was low, with 5% using emergency department services or being admitted for asthma in 2002, based on administrative data. Seventy-seven percent had 1 or more ICS dispensings during 2002, and 14% had 6 or more dispensings.

Process Results

Thirty-eight percent of intervention participants participated in the first round of calls, 32% in the second round, and 18% in the third round. Overall, 47.0% of intervention participants completed at least 1 intervention call, and 12.1% completed all 3 calls. Compared with individuals who did not participate in any of the intervention calls, members who did were older, were more likely to be female, had greater ICS usage, and reported worse asthma-specific quality of life at baseline. We did find some evidence of negative reaction to the computerized calling, in that we were able to successfully deliver the intervention more frequently with participants in the live-caller arm than in the automated-calling arm (P <.001). For instance, 59.9% of live-caller participants completed at least 1 call and 27.6% completed all 3 calls. Comparable numbers for the automated-caller arm were 46.3% and 11.2%, respectively.

Nine percent of completed intervention calls resulted in a high-priority alert and, of these, 54% resulted in some type of documented follow-up contact by the health plan. These contacts generally consisted of only telephone follow-up, although 28% of these follow-up contacts (15% of all high-priority calls) resulted in a visit being scheduled. In addition, 18% of all high-priority calls resulted in some type of medication adjustment. Among individuals who participated in the intervention, satisfaction with the calls was generally high. This finding was evidenced by the fact that 95% of calls, once started, were successfully completed, and 97% of participants who completed a call agreed to be called back 5 months later. In addition, of 269 individuals who completed a series of feedback questions on the follow-up mail survey, 81% reported that they appreciated receiving the calls, 56% indicated the calls were helpful, 14% noted that they or their provider made changes to how they cared for their asthma, and 36% said the calls made them feel differently about their asthma. These figures were even higher among the 630 participants who provided feedback immediately after the final intervention call; for example, 85% indicated that the calls were helpful to them and 24% reported that they had changed their asthma care as a result of the calls. For both the mailed survey and the post-call feedback, the response patterns were similar for participants in the live-caller and automated- calling arms of the intervention. When asked specifically what they found helpful about the intervention, participants most frequently reported increased knowledge or awareness of asthma (35% of the 161 individuals who responded to this question on the mailed questionnaire). When asked what changes to their asthma care the program brought about, half of the 58 respondents to this question indicated a medication change and one third noted that increased awareness of their asthma led to better care. Among the 82 individuals who reported that they chose not to participate in the intervention, the most frequently cited reasons were no time (38%), did not want to talk to a computer (33%), and asthma already well controlled (32%).

We also sent surveys to 236 providers whose patients completed 1 or more intervention contacts. Of 150 responders, 64% reported that the program helped with patient education regarding medications and treatment plans, 59% stated that the program helped them to clinically manage their asthma patients better, and 70% stated that this or a similar program should be continued in the future (Table 3).

Outcomes Results

The overall intent-to-treat analysis found no significant differences between the intervention and usual-care groups in terms of medication use (Table 4), healthcare utilization (Table 5), or patient-centered outcomes relevant to asthma control and quality of life (Table 6). These treatment comparisons were similarly not significant for the subset of participants younger than 65 years and for the subset of patients with more severe disease (data not shown).

Although the overall intent-to-treat analyses gave nonsignificant results, post hoc analyses that compared the control participants to participants who actually used the intervention found numerous significant, albeit small, differences. Intervention users received significantly more ICS dispensings and significantly fewer short-acting b-agonist dispensings during the 12 months after randomization, and were more likely to report better asthma-specific quality of life (emotional and symptom domains of the Asthma Quality of Life Questionnaire) and satisfaction with asthma care. They were also significantly more likely to have a routine visit for asthma during the follow-up period. Due to the previously noted differences between intervention users and nonusers, the P values for these analyses were all adjusted for age, sex, and number of ICS dispensings in 2002.

DISCUSSION

In this effectiveness study of the impact of an automated telephone outreach system to help improve asthma care, the primary intent-to-treat analysis found no significant differences between intervention and usual-care participants in terms of medication use, use of acute healthcare services, or patient-centered outcomes including asthma control, asthma-specific quality of life, and satisfaction with asthma care. These results held both overall and for subgroups defined by baseline severity and by age. We did observe statistically significant differences among the subset of intervention participants who actually participated in the intervention program, although the magnitude of these differences was small and of uncertain clinical significance. Additionally, we were able to successfully integrate the intervention into ongoing care management, and the intervention appeared to be well received by both providers and by patients who interacted with it.

Relatively few studies have assessed the utility of telephone outreach as part of an asthma management program, and those that have used live-person callers as opposed to automated telephone calls. Although manual telephone outreach may have economic benefits relative to a system of more traditional clinic visits,²² reaching patients can be very time consuming, limiting the number of patients that can be effectively managed by a given care manager. The use of automated telephone calling offers at least the promise of a time-and cost-effective means to contact thousands of individuals, thus greatly extending the reach of traditional care management services. Although we are unaware of prior use of automated telephone interventions for asthma disease management, their use has been shown to be acceptable to patients in other clinical areas and has been shown to be effective.^23-27 For example, evidence from controlled trials supports the efficacy of automated voice messaging for preventive care education and as reminders for childhood and influenza immunizations, tuberculosis control, and medication compliance.^28-32 Automated voice messaging also has proved efficacious in the management of cardiovascular disease³³ and diabetes.³⁴ Early evidence shows that automated voice messaging can be useful in improving adult eating habits³⁵ and increasing physical activity.³⁶ A recent systematic review of interactive voice response-mediated applications³⁷ concluded that they can significantly affect behavioral and clinical outcomes. Despite evidence that interactive voice response-based interventions for education and information are acceptable to patients,^37,38 reliance on pressing phone buttons for responses may decrease usability for some patients. As in our study, advanced speech-recognition software allows patients to respond more naturally to questions, thus potentially enhancing acceptance and effectiveness of this form of telephone-based interaction.^39,40 The present study differed in several important ways from previous investigations that incorporated telephone outreach for asthma management. First, the automated nature of the calls limited the "feel" and content of the calls. Although we were able to provide tailored feedback to participants, the information was necessarily educational in nature; trying to provide traditional clinical advice would have been ethically inappropriate.

Second, the intervention consisted exclusively of the calls and was not combined with, for example, clinical visits or measurements (although we did alert the patients' primary care provider if we thought the patient was at high risk for a future acute exacerbation). Third, this trial was a true effectiveness study and was not limited to a narrow, self-selected subgroup of participants. We identified a large group of patients with asthma and analyzed all of them on an intent-to-treat basis.

Fourth, the intervention was intentionally conceived of as a low-intensity, cost-effective public health-based intervention modeled after the highly successful model for public health-based approaches to tobacco cessation within the healthcare setting.⁴¹ The hope from the beginning was that we might achieve intervention effects that, even though modest, when applied to a large population would have important public health implications.

Finally, the intervention was conducted within the context of the relatively well-managed members of a large health maintenance organization that already devoted considerable energy to asthma disease management. More than three quarters of the study population had 1 or more dispensings of an ICS during the baseline year, only 5% had an emergency department visit or hospitalization for asthma during the baseline year, and only 6% were current smokers. Three quarters reported excellent, good, or very good health status, and mean asthma quality-of-life scores were high. Thus, in retrospect, there was little room for improvement in these outcomes. Nevertheless, it is plausible that the same intervention, if delivered to a less well-controlled population, might result in significant improvement in health outcomes. Additional explanations for the lack of effect relate to the lack of adoption of the intervention and the nature of the intervention itself. Only 47% of participants completed even 1 call, and only 12% completed all 3 calls. Part of the nonparticipation was related to lack of willingness to talk to a computer, but lack of time and "my asthma is already well controlled" were cited equally frequently as reasons for not participating in the calls.⁴² Even ignoring these issues, the intervention content may simply have been too low intensity and nonspecific, or may have been pitched at too broad an audience. Perhaps automated, phone-based interventions, to be effective in asthma, need to be much more focused. An example might be telephone calls specifically designed to promote medication adherence and made to patients who have not filled new prescription orders (or refilled existing prescriptions) within a prespecified time window. The World Health Organization recently concluded that, "[i]nterventions aimed at improving adherence would provide a significant positive return on investment through primary prevention (of risk factors) and secondary prevention of adverse health outcomes."⁴³ Despite the lack of significant effects on health outcomes, participants who used the service perceived value from it. Almost all agreed to receive subsequent calls, and among 630 patients who completed a brief survey immediately after the last call, 85% indicated that the calls were helpful to them, and 24% reported that they had changed their asthma care as a result of the calls. Among 150 clinicians surveyed about the program, 59% reported that the calls helped them better clinically manage their patients with asthma, and 70% indicated that the service should be continued.⁴² In summary, it might be feasible to incorporate automated telephone calling as part of an overall asthma disease management program, at least in the context of a managed care organization serving a defined population. Such a service, if carefully constructed with input from key stakeholders, can be well accepted by patients as well as providers. The main challenge appears to be getting patients to use the service in the first place. For asthma, such reluctance appears to result from lack of time, a sense that one's asthma is already well controlled, and reluctance to interact with a computer. While the present intervention did not improve health outcomes in this particular study, our study does encourage further research into interactive voice recognition-mediated interventions. Such interventions, when adopted as part of a broader disease management program, may well have a role in chronic disease management in general. With better adherence or a more focused intervention, such a service also may have a role specifically in the management of chronic asthma. Further studies of this technology seem warranted.

Acknowledgment

The authors gratefully acknowledge the editorial assistance of Ms Martha Swain in combining 2 originally submitted papers into the current single document.

Appendix: Phone Intervention Questions

1. Have you had to stay overnight in the hospital for your breathing problems in the past 3 months?

If YES to question 1, ask 1A and then skip to question 3.

1A. Have you had a chance to follow up with a Kaiser Permanente health professional about this visit?

2. Have you been treated in an emergency room for breathing problems within the past 3 months?

If YES to question 2, ask 2A.

2A. Have you had a chance to follow up with a Kaiser Permanente health professional about this visit?

[These next few questions are about your experience with your asthma in the past 4 weeks.]

3. In the past 4 weeks, did you feel that your asthma was well controlled?

4. In the past 4 weeks, did your asthma wake you up at night?

5. In the past 4 weeks, did you miss any work, school, or normal daily activity (for example, household chores or social activities) because of your asthma?

6. Do you use an inhaler for quick relief from asthma symptoms?

7. Did you take more than 12 puffs of your quick-relief inhaler on any day in the past 4 weeks?

[What medications you take and how you take them can change over time. For these next questions, we are interested in how you are currently taking your medication. Let's start with short-acting b-agonist inhalers, such as Ventolin and Albuterol, which are used for quick relief of asthma symptoms.]

8. Do you have a short-acting b-agonist inhaler?

If YES to question 8, ask 8A, otherwise skip to question 9.

8A. Do you normally use it every day or almost every day? If NO to question 8A, ask 8B, otherwise skip to question 9.

8B. Are you using it mainly when you have symptoms?

9. How about inhaled steroids, such as Qvar, Azmacort, Flovent, Pulmicort, and Advair. Do you have an inhaled steroid?

If YES to question 9, ask 9A, otherwise skip to question 10.

9A. Do you normally use it every day or almost every day? If NO to question 9A, ask 9B, otherwise skip to question 10.

9B. Are you using it mainly when you have symptoms?

10. Now I'd like to ask you about your use of long-acting b-agonist inhalers such as Serevent, Foradil, and Salmeterol. Do you have a long-acting b-agonist inhaler?

If YES to question 10, ask 10A, otherwise skip to question 11.

10A. Do you normally use it every day or almost every day? If NO to question 10A, ask 10B, otherwise skip to question 11.

10B. Are you using it mainly when you have symptoms?

11. Finally, do you have a prescription for Singulair, Montelukast, Accolate, or any other so-called leukotriene modifier?

If YES to question 11, ask 11A, otherwise skip to question 12.

11A. Do you normally use it every day or almost every day?

If NO to question 11A, ask 11B, otherwise skip to question 12.

11B. Are you using it mainly when you have symptoms?

12. Do you have a regular doctor who you see for your breathing problems?

From the Center for Health Research, Kaiser Permanente Northwest, Portland, Ore (WMV, DP, EAO, EAF, DM); Sage Marketing Associates, LLC, Portland, Ore (MK); the Eliza Corporation, Beverly, Mass (AD); Northwest Permanente, PC, Portland, Ore (TS, TH); the Care Management Institute, Oakland, Calif (GIT); the Permanente Federation, Oakland, Calif (GIT); and Oregon Health & Science University, Portland, Ore (ASB).

Funded by the Centers for Disease Control and Prevention (200-95-0953) and the Kaiser Permanente Care Management Institute.

Address correspondence to: William M. Vollmer, PhD, Senior Investigator, Center for Health Research, 3800 N Interstate Ave, Portland, OR 97227. E-mail: william.vollmer@kpchr.org.

1. Masoli M, Fabian D, Holt S, Beasley R. The global burden of asthma: executive summary of the GINA Dissemination Committee report. Allergy. 2004;59:469-478.

2. Fuhlbrigge AL, Adams RJ, Guilbert TW, et al. The burden of asthma in the United States: level and distribution are dependent on interpretation of the national asthma education and prevention program guidelines. Am J Respir Crit Care Med. 2002;166:1044-1049.

3. Beasley R. The burden of asthma with specific reference to the United States. J Allergy Clin Immunol. 2002;109(suppl):S482-S489.

4. Mayo PH, Richman J, Harris HW. Results of a program to reduce admissions for adult asthma. Ann Intern Med. 1990;112:864-871.

5. Buchner DA, Butt LT, De Stefano A, Edgren B, Suarez A, Evans RM. Effects of an asthma management program on the asthmatic member: patient-centered results of a 2-year study in a managed care organization. Am J Manag Care. 1998;4:1288-1297.

6. MacKinnon NJ, Flagstad MS, Peterson CR, Mesch-Beatty K. Disease management program for asthma: baseline assessment of resource use. Am J Health Syst Pharm. 1996;53:535-541.

7. Lieu TA, Capra AM, Quesenberry CP, Mendoza GR, Mazar M. Computer-based models to identify high-risk adults with asthma: is the glass half empty or half full? J Asthma. 1999;36:359-370.

8. Schatz M, Zeiger RS, Vollmer WM, et al. Validation of a beta-agonist long-term asthma control scale derived from computerized pharmacy data. J Allergy Clin Immunol. 2006;117:995-1000.

9. Grana J, Preston S, McDermott PD, Hanchak NA. The use of administrative data to risk-stratify asthmatic patients. Am J Med Qual. 1997;12:113-119.

10. Schatz M, Nakahiro R, Jones CH, Roth RM, Joshua A, Petitti D. Asthma population management: development and validation of a practical 3-level risk stratification scheme. Am J Manag Care. 2004;10:25-32.

11. Lieu TA, Quesenberry CP, Sorel ME, Mendoza GR, Leong AB. Computer-based models to identify high-risk children with asthma. Am J Respir Crit Care Med. 1998;157:1173-1180.

12. Schatz M, Cook EF, Joshua A, Petitti D. Risk factors for asthma hospitalizations in a managed care organization: development of a clinical prediction rule. Am J Manag Care. 2003;9:538-547.

13. Vollmer WM, Markson LE, O'Connor EA, Frazier EA, Berger M, Buist AS. Association of asthma control with health care utilization. Am J Respir Crit Care Med. 2002;165:195-199.

14. Peters D, Chen C, Markson LE, Allen-Ramey FC, Vollmer WM. Using an asthma control questionnaire and administrative data to predict health-care utilization. Chest. 2006;129:918-924.

15. Vollmer WM, Markson LE, O'Connor EA, et al. Association of asthma control with health care utilization and quality of life. Am J Respir Crit Care Med. 1999;160:1647-1652.

16. Corkrey R, Parkinson L. Interactive voice response: review of studies 1989- 2000. Behav Res Methods Instrum Comput. 2002;34:342-353.

17. Mouza AM. IVR and administrative operations in healthcare and hospitals. J Healthc Inf Manag. 2003;17:68-71.

18. Lavigne JV, Tapper K. Interactive voice response in disease management: providing patient outreach and improving outcomes. Disease Management & Health Outcomes. 1998;4:1-16.

19. MacLean WE Jr, Perrin JM, Gortmaker S, Pierre CB. Psychological adjustment of children with asthma: effects of illness severity and recent stressful life events. J Pediatr Psychol. 1992;17:159-171.

20. Freeborn DK, Pope CR. Promise and Performance in Managed Care. 1st ed. Baltimore, Md: The Johns Hopkins University Press; 1994.

21. Juniper EF, Guyatt GH, Cox FM, Ferrie PJ, King DR. Development and validation of the mini-asthma quality of life questionnaire. Eur Respir J. 1999;14:32-38.

22. Pinnock H, Bawden R, Proctor S, et al. Accessibility, acceptability, and effectiveness in primary care of routine telephone review of asthma: pragmatic, randomised controlled trial. BMJ. 2003;326:477-479.

23. Wagner EH, Austin BT, Von Korff M. Organizing care for patients with chronic illness. Milbank Q. 1996;74:511-544.

24. Shiffman RN, Liaw Y, Brandt CA, Corb GJ. Computer-based guideline implementation systems: a systematic review of functionality and effectiveness. J Am Med Inform Assoc. 1999;6:104-114.

25. Powell MF, Burkhart VD, Lamy PP. Diabetic patient compliance as a function of patient counseling. Drug Intell Clin Pharm. 1979;13:506-511.

26. Balas EA, Boren SA, Griffing G. Computerized management of diabetes: a synthesis of controlled trials. Proc AMIA Symp. 1998:295-299.

27. Rich MW, Gray DB, Beckham V, Wittenberg C, Luther P. Effect of a multidisciplinary intervention on medication compliance in elderly patients with congestive heart failure. Am J Med. 1996;101:270-276.

28. Boult C, Pualwan TF, Fox PD, Pacala JT. Identification and assessment of highrisk seniors. HMO Workgroup on Care Management. Am J Manag Care. 1998;4:1137-1146.

29. Blalock SJ, Currey SS, DeVellis RF, et al. Effects of educational materials concerning osteoporosis on women's knowledge, beliefs, and behavior. Am J Health Promot. 2000;14:161-169.

30. Wagner TH. The effectiveness of mailed patient reminders on mammography screening: a meta-analysis. Am J Prev Med. 1998;14:64-70.

31. Marci CD, Anderson WB, Viechnicki MB, Greenspan SL. Bone mineral densitometry substantially influences health-related behaviors of postmenopausal women. Calcif Tissue Int. 2000;66:113-118.

32. DeBusk RF, Miller NH, Superko HR, et al. A case-management system for coronary risk factor modification after acute myocardial infarction. Ann Intern Med. 1994;120:721-729.

33. Kanis JA, Melton LJ, III, Christiansen C, Johnston CC, Khaltaev N. Perspective: the diagnosis of osteoporosis. J Bone Miner Res. 1994;9:1137-1141.

34. Weinberger M, Kirkman MS, Samsa GP, et al. A nurse-coordinated intervention for primary care patients with non-insulin-dependent diabetes mellitus: impact on glycemic control and health-related quality of life. J Gen Intern Med. 1995;10:59-66.

35. Bates DW. Using information technology to reduce rates of medication errors in hospitals. BMJ. 2000;320:788-791.

36. Heffner JE. The overarching challenge. Chest. 2000;118(suppl):1S-3S.

37. Krishna S, Balas EA, Boren SA, Maglaveras N. Patient acceptance of educational voice messages: a review of controlled clinical studies. Methods Inf Med. 2002;41:360-369.

38. Glasgow RE, McKay HG, Piette JD, Reynolds KD. The RE-AIM framework for evaluating interventions: what can it tell us about approaches to chronic illness management? Patient Educ Couns. 2001;44:119-127.

39. Boyce C. Natural spoken dialogue system for telephony applications. Communications of the ACM. 2000;43:29-34.

40. Suhm B, Bers J, McCarthy D, et al. A comparative study of speech in the call center: natural language call routing vs. touch-tone menus. Presented at: Conference on Human Factors in Computing Systems: Changing Our World, Changing Ourselves. Minneapolis, Minn; April 20-25, 2002:283-290.

41. Meenan R, Stevens V, Hornbrook M, et al. Cost-effectiveness of a hospital-based smoking cessation intervention. Med Care. 1998;36:670-678.

42. DeBusk RF, Miller NH, Superko HR, et al. A case-management system for coronary risk factor modification after acute myocardial infarction. Ann Intern Med. 1994;120:721-729.

43. World Health Organization. Adherence to long-term therapies: evidence for action. Noncommunicable diseases and mental health adherence to long-term therapies. Geneva, Switzerland: World Health Organization; January 2003. Available at: http://www.who.int/chronic_conditions/adherencereport/en/. Accessed September 19, 2006.

Download PDF: Use and Impact of an Automated Telephone Outreach System for Asthma in a Managed Care Setting