Telemonitoring Heart Failure Patients -- Any Benefit?
Telemonitoring Heart Failure Patients -- Any Benefit?
Chaudhry SI, Mattera JA, Curtis JP, et al. Telemonitoring in patients with heart failure. N Engl J Med. 2010;363:2301-2309.
This study was chosen as the subject of an interview because of its selection in Medscape Best Evidence. This service ranks studies in 2 ways: by analyzing their statistical strength using a 15-item rating instrument and by the assessment of clinical relevance as determined by an expert review panel.
Dr. Chaudhry is Assistant Professor of Medicine in the Department of Internal Medicine at Yale University School of Medicine, New Haven, Connecticut. Her research interests include cardiovascular outcomes, geriatric conditions, disability, quality of life, remote monitoring, and end of life issues. Dr. Chaudhry's research is directed toward improving outcomes in older patients with cardiovascular disease. She is currently funded by National Institutes of Health (NIH)/National Institute on Aging (NIA) to examine the prognostic importance of geriatric impairments in heart failure ,including impairments in cognition, vision, hearing, muscle strength, and physical capacity. She was Director of Science of the Telemonitoring to Improve Heart Failure Outcomes (Tele-HF) study.
Identifying successful strategies to prevent acute decompensation and/or hospital readmission in stable patients with chronic heart failure continues to challenge researchers. Telemonitoring (ie, transfer of data from patients at home to their healthcare provider) has been suggested as a way of frequently assessing a patient's heart failure status that would result in earlier recognition of clinical deterioration compared with usual clinical practice. The benefits of telemonitoring on heart failure outcomes appeared to have been suggested in small studies and by meta-analyses of data from these studies. Until recently, however, the benefits have not been established in a large trial. The Tele-HF study was carried out to determine whether automated symptom and self-reported weight monitoring would reduce a combined endpoint of all-cause hospitalization and mortality in patients recently hospitalized for heart failure. This large, multicenter US study, which was carried out between 2006 and 2009, was supported by the National Heart, Lung and Blood Institute (NHLBI). The results were presented at the 2010 Scientific Sessions of the American Heart Association and published simultaneously in the New England Journal of Medicine. Unexpectedly, they failed to show any benefit associated with telemonitoring compared with usual care.
The Tele-HF study enrolled patients with a history of heart failure admission within 30 days at 33 US cardiology practices. A total of 1653 patients, median age 61 years, 42.0% female, were randomly assigned to undergo either telemonitoring (826 patients) or usual care (827 patients). Baseline characteristics were similar in the 2 groups. Usual care consisted of treatment in accordance with the American College of Cardiology/American Heart Association guidelines for the management of heart failure. Telemonitoring was accomplished by means of a commercial telephone-based interactive voice response system, Tel-Assurance™ (Pharos Innovations, Chicago, IL). Patients received education and instruction about the system at the time of enrollment. They made free calls to the system daily and responded using the telephone keypad to a series of questions about weight, symptoms, and general health. These responses were uploaded to a secure Website and reviewed every weekday by clinicians at each practice site. The system was preprogrammed to alert clinicians in the event of certain responses (ie, those denoting possible clinical deterioration) and in those cases, calls were made to each patients to obtain more information and offer advice. If a patient did not call the system for 2 consecutive days, they received a system-generated reminder call and a call from site staff.
In both groups, 79% of patients completed the final, 6-month follow-up interview. There was no significant difference between the telemonitoring and usual-care groups with respect to the primary endpoint, readmission for any reason, or death from any cause within 180 days after enrollment, which occurred in 52.3% and 51.5% of patients, respectively. Nor were significant differences seen between the 2 groups with respect to the secondary endpoints, including readmission for any reason (49.3% vs 47.4%) and death (11.1% vs 11.4%), as well as readmissions for heart failure, number of days in hospital, and number of admissions. Subgroup analyses including age, gender, race, left ventricular ejection fraction (LVEF), and New York Heart Association (NYHA)functional class did not identify any patient characteristic associated with greater benefit from telemonitoring. The telemonitoring intervention did not appear any more beneficial at more experienced sites (ie, those that enrolled ≥ 100 patients).
In their published report, Dr. Chaudhry and colleagues pointed out although the results contradicted the findings of the Cochrane meta-analysis, the Tele-HF trial was larger and "of higher methodological quality" than most of the studies included in the meta-analysis, which raised questions about the value of its findings. They emphasized the importance of "a thorough, independent evaluation" of a disease-management strategy before it is widely adopted. They acknowledged that their findings indicated a need to examine alternative approaches to improving care. In an accompanying editorial, Akshay S. Desai, MD and Lynne Warner Stevenson, MD (Brigham and Women's Hospital, Boston) concluded that although the "unequivocal message" from the study was that this application of telemonitoring did not improve outcomes after hospital discharge, it did not preclude its potential role as a useful adjunct to disease management. "Heart failure is a dynamic disease that will most likely require some form of surveillance in the home after hospital discharge in order to reduce the nearly 50% rate of hospital readmission at 6 months seen the study," they pointed out.
Although the Tele-HF results challenged the findings of previous studies, another large, multicenter, randomized trial of telemonitoring, presented at the same the American Heart Association meeting, also failed to demonstrate a benefit of a telemonitoring system. In the Telemedical Interventional Monitoring in Heart Failure (TIM-HF) study, researchers at 165 sites in Germany enrolled a total of 710 patients with NYHA class II/III heart failure and LVEF ≤ 35%, with a history of cardiac decompensation with hospitalization in the past 18 months. Patients were randomly assigned to an intervention with daily remote device monitoring (ECG, blood pressure, body weight) coupled with medical telephone support or to usual care led by the patients' local physician. The results showed no difference in hospitalization or survival rates between the 2 arms of the study, although subgroup analysis suggested potential benefit in patients with prior heart failure hospitalization and EF > 25%. Based on the Tele-HF and TIM-HF studies and other evidence, a recent review of telemonitoring in chronic heart failure concluded that "we do not know whether remote monitoring provides the best direction in managing ambulatory heart failure patients or whether it is simply a tool to identify suboptimal medical treatment."
For this interview, Dr. Chaudhry, Study Director of Tele-HF, spoke with Linda Brookes, MSc, for Medscape Cardiology, about the clinical implications of the Tele-HF study for improving heart failure outcomes using telemonitoring intervention.
Medscape: In your report you made the point that Tele-HF was a more rigorous study, and it was also larger than previous studies of telemonitoring in heart failure patients. So can we take the results as conclusive evidence that telemonitoring isn't beneficial in this group?
Sarwat I. Chaudhry, MD: It is important to remember when people talk about telemonitoring, that it can take many different forms, and ours was a very rigorous evaluation of 1 particular form of telemonitoring. Therefore, keep in mind that the findings of our study were not indications that all of telemonitoring is of no benefit for heart failure patients. It may be that other forms of telemonitoring could prove more helpful. Ours was a study in which patients made calls into an automated telemonitoring system. However, if there were any concerning clinical responses then contact was made with a clinician, so it was not entirely automated. Other forms of telemonitoring include patients having regular ongoing home visits with a visiting nurse or ongoing telephone calls with nurses, pharmacists -- a sort of multidisciplinary team approach. So it may be that an approach that involves more human contact may be more beneficial. It is also possible that an approach that targets not just increased monitoring, but also that critical transition from hospital to home may prove more beneficial. So there are many different ways that the intervention itself could be altered and potentially prove to be more effective.
The approach we tested involved providing doctors with more information about their patients' clinical status. The questions that heart failure patients were asked every day in this automated monitoring system were really the same questions that we would ask these patients if they came into our office: how were they feeling generally? Had they noticed more swelling in the last day? What was their weight that morning? How were their energy levels? That sort of thing. Then we relayed that information to clinicians in a way that allowed them to review it very efficiently.
It was an approach that made a lot of clinical sense, and we had great faith that it would work. Smaller studies had proven promising, and we though that perhaps we would see the same result in this larger study, so we were somewhat surprised at these results. It speaks to the importance of a rigorous evaluation of any approach, even one that seems to make good clinical sense, because before we can put our resources into such intensive systems, we need to have better evidence that they really are going to improve outcomes.
Medscape: Given the disappointing results, do you think anything could have been done differently to make telemonitoring more beneficial in the study, for example, if the study had been longer or if it had involved more education?
Dr. Chaudhry: You know, I would have to say I don't think so. We really optimized the intervention, the condition, and the setting that we were testing it. In fact, if the study had been positive, we would have had to think carefully about generalizability, because we really handpicked the sites and the clinical practices that participated. They had very motivated doctors, very motivated nurses, and very motivated patients. Many patients were screened for inclusion in the study and declined using it right away or after hearing about the study said, "No, thank you." So we ended up with a group that was quite motivated. We worked very closely and intensively with the practice sites. We went out and visited them and we had ongoing support 24 hours per day. We had monthly calls with all of the participating nurses and physicians to maintain enthusiasm and to troubleshoot ways to use this efficiently.
As far as patient education was concerned, I cannot really think of a way that would have been more effective. All patients in the study received educational handouts developed specifically by the Heart Failure Society of America for taking care of their heart failure, and patients had frequent contact with clinicians. Any time there was an abnormal response to the telemonitoring -- for example, if a patient said "I get short of breath" or "my weight went up" -- they got a phone call, usually from a nurse, who would talk through what their diet was, how they were taking their medications, and so forth. So there were many educational interventions happening. We solved problems in the patient self-reports, for example, worrying about eating a smoked turkey sandwich or not taking medications. The nurses talked with patients in great detail and counseled them, enforcing the need to adhere to an appropriate diet and explaining what that is, supporting medications, and troubleshooting any barriers to all of those things. So it is hard for us to think that small changes might have changed the results, because really we tried to optimize everything with respect to this form of intervention, but again, whether a different form of intervention might be more effective remains an open question.
Medscape: Subgroup analyses of the data presented from the TIM-HF study suggested that telemonitoring might be beneficial in patients with prior heart failure hospitalization and an EF > 25%. In all the analyses of the Tele-HF data that you have done, did you find any subgroup, however small, that might have benefited from telemonitoring in the study?
Dr. Chaudhry: No. We have looked at the data in many different ways, but we did not find any evidence of benefit in subgroup analyses.
Medscape: In your paper you reported that about 14% of patients randomly assigned to undergo telemonitoring never actually used it. In addition, by the final week of the study period, only 55% of the patients were still using the system a minimum of 3 times per week. The adherence rates in this trial probably represent the best-case scenario. So if this is the rate in a highly motivated group of patients working with highly motivated staff, the message seems to be not just that telemonitoring is not effective, but that it will be really hard to get people to use any kind of system like this.
Dr. Chaudhry: That is right. I think that is exactly the message. It is not surprising, given what we know about medication adherence. Any time we ask patients to do something on a daily basis, whether it be to take a medication or alter their diet or increase their physical activity, it is difficult for many patients to implement.
Medscape: It is understood that adherence is a challenge in patients with prescriptions for treating hypertension or hypercholesterolemia, because there aren't any symptoms that the patient can recognize. But heart failure patients have easily recognizable symptoms like shortness of breath and swelling up with sudden weight gain, so wouldn't these patients better understand the need to adhere to any intervention?
Dr. Chaudhry: We did talk to a number of patients about their experiences using this system, and we heard a lot of different reasons why it was difficult to keep using it. Some patients were younger folks who were still working and it was difficult to fit it into their busy schedules, especially because patients had to weigh themselves in the morning and then make the phone call. Other patients felt that they were feeling healthy, and they did not want to be bothered with confirming to a system that they were feeling fine. On the other hand, some patients were feeling too ill to use the system. So there was a whole spectrum of reasons why people didn't use it.
Medscape: If you looked only at the people who did use the system, would you see a beneficial effect in that group?
Dr Chaudhry: Of course that is a really difficult question to answer, because we know that people who are adherent generally do better. So we have not looked at the data in that way, and we are not really convinced that we would see something that would be meaningful, because, again, patients who do as they are instructed -- taking a medication, exercising, or whatever it is -- generally do better, whether it is because they are healthy enough to follow instructions or they have a social situation that is organized and structured to enable them to follow such instructions. However, adherence was certainly an ongoing challenge. In this clinical trial, we had a number of resources available to maximize adherence. So patients who did not make a phone call got a call immediately from patient engagers who were employed by the telemonitoring company specifically to make such calls and follow-up, to go through the benefits and address failures and work with the patients. So they had a number of enablers to help maximize use of the system and yet an adherence rate of only 55% was still what we were able to achieve.
Medscape: Would you be publishing any follow-up to this about the reasons why you think that telemonitoring wasn't successful in this study?
Dr Chaudhry: Yes, we do have some qualitative ongoing work addressing adherence, and that work will be reported.
Medscape: The authors of the New England Journal of Medicine editorial questioned whether the signals of weight and symptoms, as listed on the questionnaire, provided adequate warning of the onset of decompensation. They also suggested that the team member receiving the data should be a midlevel specialist in heart failure management, empowered to contact the patient directly with a treatment plan without having to "triangulate" the discussion with a physician before recommending a plan. Do you think making such changes might have made a difference to the results?
Dr. Chaudhry: We actually did have direct communication with a midlevel person who often was enabled to make changes and contact the patient directly. Although the editorial suggested that there was a delay, because the person calling the patient could not make any changes to the treatment that was not the case. The person who was contacting the patient was often a midlevel provider who could alter or schedule an appointment, educate about diet, or adjust medications, so there wasn't any additional delay.
Some people suspect that symptoms and body weight are not adequately sensitive and that perhaps with more invasive physiological measurement, such as intracardiac pressure, you might get an earlier signal of decompensation. We do not know that, but it could be tested. Certainly the questions asked in the system were those we ask patients when they present to the hospital as routine standard of care, so I don't know any different symptom questions that we could have added. However, the one thing that might be questioned is whether more physiologic measurements might have proved more effective.
Medscape: The TIM-HF study reported the same lack of benefit with a different telemonitoring system. Is it really possible that another system could affect outcomes positively?
Dr. Chaudhry: I think we can say clearly that automated telemonitoring, and monitoring of symptoms and weight does not work.
Medscape: So do you think any studies are needed or will be done using a different approach to monitoring?
Dr Chaudhry: I think, based on what I see and know, that there are a lot of challenges in transitioning sick patients in the hospital back to home, and that we need to look carefully at that transition. This is something that our group is actively involved with.. There are a number of challenges. However, simply providing physicians with more information about their patients' clinical status does not work. We are going to need more complex interventions to address that transition from hospital to home, looking at all the patients' critical vulnerabilities, whether they be physical, cognitive, or social, that may be barriers to their ability to obtain medications and take them every day and to take care of themselves in the way that we envision them doing. We need to really be sure that the patients understand what changes in their health they should be looking out for and when they start to experience decompensations that they know whom to contact in a timely fashion.
Medscape: What kind of vulnerabilities would you be looking at?
Dr. Chaudhry: Many of our patients are physically unable to get to the pharmacy and get their medication, something as simple as that. When they get the medications their vision may be so poor that they are unable to completely read the directions. They may be so weak that they are unable to stand up on a scale to measure their body weight. Patients with cognitive decline are unable to remember to take their medication. These are simple things that sometimes get in the way of patients being able to take care of themselves in the way that we envision they should.
Medscape: So how are you following up this study? How do you follow up Tele-HF?
Dr. Chaudhry: That is something we are still discussing. We are looking at novel ways of reducing readmission rates, given that this approach did not work. I think we are going to need a more multidisciplinary holistic approach, one that incorporates care transitions and patient vulnerabilities to effectively deliver a program that is feasible for patients and clinicians. Of course, economically that is a challenge. That is something we are actively planning for now.
Medscape: You mentioned economics. In terms of additional education and personnel that must be a big consideration.
Dr. Chaudhry: I think it makes a compelling case. It is imperative that we continue to study this problem of readmissions, because it is very bothersome and detrimental for patients and for their families, and from the societal perspective it is crucial that we come up with new strategies that will prove more effective.
Medscape: Are you and your colleagues working exclusively with heart failure patients or are you working in other specialties?
Dr. Chaudhry: I am a general internist by training, so I am very interested in this issue of hospitalizations and readmissions for patients with a number of different diseases, including myocardial infarction and community-acquired pneumonia.
Medscape: Are the data relevant from one specialty to another?
Dr. Chaudhry: There are some relevant messages. I think particularly the issue of the complexity of patients and the challenges of transitioning from hospital to home are relevant themes for patients with a number of different diagnoses, including heart failure.
A Best Evidence Interview with Sarwat I. Chaudhry, MD
The Best Evidence Study
Chaudhry SI, Mattera JA, Curtis JP, et al. Telemonitoring in patients with heart failure. N Engl J Med. 2010;363:2301-2309.
This study was chosen as the subject of an interview because of its selection in Medscape Best Evidence. This service ranks studies in 2 ways: by analyzing their statistical strength using a 15-item rating instrument and by the assessment of clinical relevance as determined by an expert review panel.
About the Interviewee
Dr. Chaudhry is Assistant Professor of Medicine in the Department of Internal Medicine at Yale University School of Medicine, New Haven, Connecticut. Her research interests include cardiovascular outcomes, geriatric conditions, disability, quality of life, remote monitoring, and end of life issues. Dr. Chaudhry's research is directed toward improving outcomes in older patients with cardiovascular disease. She is currently funded by National Institutes of Health (NIH)/National Institute on Aging (NIA) to examine the prognostic importance of geriatric impairments in heart failure ,including impairments in cognition, vision, hearing, muscle strength, and physical capacity. She was Director of Science of the Telemonitoring to Improve Heart Failure Outcomes (Tele-HF) study.
Introduction to the Interview
Identifying successful strategies to prevent acute decompensation and/or hospital readmission in stable patients with chronic heart failure continues to challenge researchers. Telemonitoring (ie, transfer of data from patients at home to their healthcare provider) has been suggested as a way of frequently assessing a patient's heart failure status that would result in earlier recognition of clinical deterioration compared with usual clinical practice. The benefits of telemonitoring on heart failure outcomes appeared to have been suggested in small studies and by meta-analyses of data from these studies. Until recently, however, the benefits have not been established in a large trial. The Tele-HF study was carried out to determine whether automated symptom and self-reported weight monitoring would reduce a combined endpoint of all-cause hospitalization and mortality in patients recently hospitalized for heart failure. This large, multicenter US study, which was carried out between 2006 and 2009, was supported by the National Heart, Lung and Blood Institute (NHLBI). The results were presented at the 2010 Scientific Sessions of the American Heart Association and published simultaneously in the New England Journal of Medicine. Unexpectedly, they failed to show any benefit associated with telemonitoring compared with usual care.
The Tele-HF study enrolled patients with a history of heart failure admission within 30 days at 33 US cardiology practices. A total of 1653 patients, median age 61 years, 42.0% female, were randomly assigned to undergo either telemonitoring (826 patients) or usual care (827 patients). Baseline characteristics were similar in the 2 groups. Usual care consisted of treatment in accordance with the American College of Cardiology/American Heart Association guidelines for the management of heart failure. Telemonitoring was accomplished by means of a commercial telephone-based interactive voice response system, Tel-Assurance™ (Pharos Innovations, Chicago, IL). Patients received education and instruction about the system at the time of enrollment. They made free calls to the system daily and responded using the telephone keypad to a series of questions about weight, symptoms, and general health. These responses were uploaded to a secure Website and reviewed every weekday by clinicians at each practice site. The system was preprogrammed to alert clinicians in the event of certain responses (ie, those denoting possible clinical deterioration) and in those cases, calls were made to each patients to obtain more information and offer advice. If a patient did not call the system for 2 consecutive days, they received a system-generated reminder call and a call from site staff.
In both groups, 79% of patients completed the final, 6-month follow-up interview. There was no significant difference between the telemonitoring and usual-care groups with respect to the primary endpoint, readmission for any reason, or death from any cause within 180 days after enrollment, which occurred in 52.3% and 51.5% of patients, respectively. Nor were significant differences seen between the 2 groups with respect to the secondary endpoints, including readmission for any reason (49.3% vs 47.4%) and death (11.1% vs 11.4%), as well as readmissions for heart failure, number of days in hospital, and number of admissions. Subgroup analyses including age, gender, race, left ventricular ejection fraction (LVEF), and New York Heart Association (NYHA)functional class did not identify any patient characteristic associated with greater benefit from telemonitoring. The telemonitoring intervention did not appear any more beneficial at more experienced sites (ie, those that enrolled ≥ 100 patients).
In their published report, Dr. Chaudhry and colleagues pointed out although the results contradicted the findings of the Cochrane meta-analysis, the Tele-HF trial was larger and "of higher methodological quality" than most of the studies included in the meta-analysis, which raised questions about the value of its findings. They emphasized the importance of "a thorough, independent evaluation" of a disease-management strategy before it is widely adopted. They acknowledged that their findings indicated a need to examine alternative approaches to improving care. In an accompanying editorial, Akshay S. Desai, MD and Lynne Warner Stevenson, MD (Brigham and Women's Hospital, Boston) concluded that although the "unequivocal message" from the study was that this application of telemonitoring did not improve outcomes after hospital discharge, it did not preclude its potential role as a useful adjunct to disease management. "Heart failure is a dynamic disease that will most likely require some form of surveillance in the home after hospital discharge in order to reduce the nearly 50% rate of hospital readmission at 6 months seen the study," they pointed out.
Although the Tele-HF results challenged the findings of previous studies, another large, multicenter, randomized trial of telemonitoring, presented at the same the American Heart Association meeting, also failed to demonstrate a benefit of a telemonitoring system. In the Telemedical Interventional Monitoring in Heart Failure (TIM-HF) study, researchers at 165 sites in Germany enrolled a total of 710 patients with NYHA class II/III heart failure and LVEF ≤ 35%, with a history of cardiac decompensation with hospitalization in the past 18 months. Patients were randomly assigned to an intervention with daily remote device monitoring (ECG, blood pressure, body weight) coupled with medical telephone support or to usual care led by the patients' local physician. The results showed no difference in hospitalization or survival rates between the 2 arms of the study, although subgroup analysis suggested potential benefit in patients with prior heart failure hospitalization and EF > 25%. Based on the Tele-HF and TIM-HF studies and other evidence, a recent review of telemonitoring in chronic heart failure concluded that "we do not know whether remote monitoring provides the best direction in managing ambulatory heart failure patients or whether it is simply a tool to identify suboptimal medical treatment."
For this interview, Dr. Chaudhry, Study Director of Tele-HF, spoke with Linda Brookes, MSc, for Medscape Cardiology, about the clinical implications of the Tele-HF study for improving heart failure outcomes using telemonitoring intervention.
The Interview
Medscape: In your report you made the point that Tele-HF was a more rigorous study, and it was also larger than previous studies of telemonitoring in heart failure patients. So can we take the results as conclusive evidence that telemonitoring isn't beneficial in this group?
Sarwat I. Chaudhry, MD: It is important to remember when people talk about telemonitoring, that it can take many different forms, and ours was a very rigorous evaluation of 1 particular form of telemonitoring. Therefore, keep in mind that the findings of our study were not indications that all of telemonitoring is of no benefit for heart failure patients. It may be that other forms of telemonitoring could prove more helpful. Ours was a study in which patients made calls into an automated telemonitoring system. However, if there were any concerning clinical responses then contact was made with a clinician, so it was not entirely automated. Other forms of telemonitoring include patients having regular ongoing home visits with a visiting nurse or ongoing telephone calls with nurses, pharmacists -- a sort of multidisciplinary team approach. So it may be that an approach that involves more human contact may be more beneficial. It is also possible that an approach that targets not just increased monitoring, but also that critical transition from hospital to home may prove more beneficial. So there are many different ways that the intervention itself could be altered and potentially prove to be more effective.
The approach we tested involved providing doctors with more information about their patients' clinical status. The questions that heart failure patients were asked every day in this automated monitoring system were really the same questions that we would ask these patients if they came into our office: how were they feeling generally? Had they noticed more swelling in the last day? What was their weight that morning? How were their energy levels? That sort of thing. Then we relayed that information to clinicians in a way that allowed them to review it very efficiently.
It was an approach that made a lot of clinical sense, and we had great faith that it would work. Smaller studies had proven promising, and we though that perhaps we would see the same result in this larger study, so we were somewhat surprised at these results. It speaks to the importance of a rigorous evaluation of any approach, even one that seems to make good clinical sense, because before we can put our resources into such intensive systems, we need to have better evidence that they really are going to improve outcomes.
Medscape: Given the disappointing results, do you think anything could have been done differently to make telemonitoring more beneficial in the study, for example, if the study had been longer or if it had involved more education?
Dr. Chaudhry: You know, I would have to say I don't think so. We really optimized the intervention, the condition, and the setting that we were testing it. In fact, if the study had been positive, we would have had to think carefully about generalizability, because we really handpicked the sites and the clinical practices that participated. They had very motivated doctors, very motivated nurses, and very motivated patients. Many patients were screened for inclusion in the study and declined using it right away or after hearing about the study said, "No, thank you." So we ended up with a group that was quite motivated. We worked very closely and intensively with the practice sites. We went out and visited them and we had ongoing support 24 hours per day. We had monthly calls with all of the participating nurses and physicians to maintain enthusiasm and to troubleshoot ways to use this efficiently.
As far as patient education was concerned, I cannot really think of a way that would have been more effective. All patients in the study received educational handouts developed specifically by the Heart Failure Society of America for taking care of their heart failure, and patients had frequent contact with clinicians. Any time there was an abnormal response to the telemonitoring -- for example, if a patient said "I get short of breath" or "my weight went up" -- they got a phone call, usually from a nurse, who would talk through what their diet was, how they were taking their medications, and so forth. So there were many educational interventions happening. We solved problems in the patient self-reports, for example, worrying about eating a smoked turkey sandwich or not taking medications. The nurses talked with patients in great detail and counseled them, enforcing the need to adhere to an appropriate diet and explaining what that is, supporting medications, and troubleshooting any barriers to all of those things. So it is hard for us to think that small changes might have changed the results, because really we tried to optimize everything with respect to this form of intervention, but again, whether a different form of intervention might be more effective remains an open question.
Medscape: Subgroup analyses of the data presented from the TIM-HF study suggested that telemonitoring might be beneficial in patients with prior heart failure hospitalization and an EF > 25%. In all the analyses of the Tele-HF data that you have done, did you find any subgroup, however small, that might have benefited from telemonitoring in the study?
Dr. Chaudhry: No. We have looked at the data in many different ways, but we did not find any evidence of benefit in subgroup analyses.
Medscape: In your paper you reported that about 14% of patients randomly assigned to undergo telemonitoring never actually used it. In addition, by the final week of the study period, only 55% of the patients were still using the system a minimum of 3 times per week. The adherence rates in this trial probably represent the best-case scenario. So if this is the rate in a highly motivated group of patients working with highly motivated staff, the message seems to be not just that telemonitoring is not effective, but that it will be really hard to get people to use any kind of system like this.
Dr. Chaudhry: That is right. I think that is exactly the message. It is not surprising, given what we know about medication adherence. Any time we ask patients to do something on a daily basis, whether it be to take a medication or alter their diet or increase their physical activity, it is difficult for many patients to implement.
Medscape: It is understood that adherence is a challenge in patients with prescriptions for treating hypertension or hypercholesterolemia, because there aren't any symptoms that the patient can recognize. But heart failure patients have easily recognizable symptoms like shortness of breath and swelling up with sudden weight gain, so wouldn't these patients better understand the need to adhere to any intervention?
Dr. Chaudhry: We did talk to a number of patients about their experiences using this system, and we heard a lot of different reasons why it was difficult to keep using it. Some patients were younger folks who were still working and it was difficult to fit it into their busy schedules, especially because patients had to weigh themselves in the morning and then make the phone call. Other patients felt that they were feeling healthy, and they did not want to be bothered with confirming to a system that they were feeling fine. On the other hand, some patients were feeling too ill to use the system. So there was a whole spectrum of reasons why people didn't use it.
Medscape: If you looked only at the people who did use the system, would you see a beneficial effect in that group?
Dr Chaudhry: Of course that is a really difficult question to answer, because we know that people who are adherent generally do better. So we have not looked at the data in that way, and we are not really convinced that we would see something that would be meaningful, because, again, patients who do as they are instructed -- taking a medication, exercising, or whatever it is -- generally do better, whether it is because they are healthy enough to follow instructions or they have a social situation that is organized and structured to enable them to follow such instructions. However, adherence was certainly an ongoing challenge. In this clinical trial, we had a number of resources available to maximize adherence. So patients who did not make a phone call got a call immediately from patient engagers who were employed by the telemonitoring company specifically to make such calls and follow-up, to go through the benefits and address failures and work with the patients. So they had a number of enablers to help maximize use of the system and yet an adherence rate of only 55% was still what we were able to achieve.
Medscape: Would you be publishing any follow-up to this about the reasons why you think that telemonitoring wasn't successful in this study?
Dr Chaudhry: Yes, we do have some qualitative ongoing work addressing adherence, and that work will be reported.
Medscape: The authors of the New England Journal of Medicine editorial questioned whether the signals of weight and symptoms, as listed on the questionnaire, provided adequate warning of the onset of decompensation. They also suggested that the team member receiving the data should be a midlevel specialist in heart failure management, empowered to contact the patient directly with a treatment plan without having to "triangulate" the discussion with a physician before recommending a plan. Do you think making such changes might have made a difference to the results?
Dr. Chaudhry: We actually did have direct communication with a midlevel person who often was enabled to make changes and contact the patient directly. Although the editorial suggested that there was a delay, because the person calling the patient could not make any changes to the treatment that was not the case. The person who was contacting the patient was often a midlevel provider who could alter or schedule an appointment, educate about diet, or adjust medications, so there wasn't any additional delay.
Some people suspect that symptoms and body weight are not adequately sensitive and that perhaps with more invasive physiological measurement, such as intracardiac pressure, you might get an earlier signal of decompensation. We do not know that, but it could be tested. Certainly the questions asked in the system were those we ask patients when they present to the hospital as routine standard of care, so I don't know any different symptom questions that we could have added. However, the one thing that might be questioned is whether more physiologic measurements might have proved more effective.
Medscape: The TIM-HF study reported the same lack of benefit with a different telemonitoring system. Is it really possible that another system could affect outcomes positively?
Dr. Chaudhry: I think we can say clearly that automated telemonitoring, and monitoring of symptoms and weight does not work.
Medscape: So do you think any studies are needed or will be done using a different approach to monitoring?
Dr Chaudhry: I think, based on what I see and know, that there are a lot of challenges in transitioning sick patients in the hospital back to home, and that we need to look carefully at that transition. This is something that our group is actively involved with.. There are a number of challenges. However, simply providing physicians with more information about their patients' clinical status does not work. We are going to need more complex interventions to address that transition from hospital to home, looking at all the patients' critical vulnerabilities, whether they be physical, cognitive, or social, that may be barriers to their ability to obtain medications and take them every day and to take care of themselves in the way that we envision them doing. We need to really be sure that the patients understand what changes in their health they should be looking out for and when they start to experience decompensations that they know whom to contact in a timely fashion.
Medscape: What kind of vulnerabilities would you be looking at?
Dr. Chaudhry: Many of our patients are physically unable to get to the pharmacy and get their medication, something as simple as that. When they get the medications their vision may be so poor that they are unable to completely read the directions. They may be so weak that they are unable to stand up on a scale to measure their body weight. Patients with cognitive decline are unable to remember to take their medication. These are simple things that sometimes get in the way of patients being able to take care of themselves in the way that we envision they should.
Medscape: So how are you following up this study? How do you follow up Tele-HF?
Dr. Chaudhry: That is something we are still discussing. We are looking at novel ways of reducing readmission rates, given that this approach did not work. I think we are going to need a more multidisciplinary holistic approach, one that incorporates care transitions and patient vulnerabilities to effectively deliver a program that is feasible for patients and clinicians. Of course, economically that is a challenge. That is something we are actively planning for now.
Medscape: You mentioned economics. In terms of additional education and personnel that must be a big consideration.
Dr. Chaudhry: I think it makes a compelling case. It is imperative that we continue to study this problem of readmissions, because it is very bothersome and detrimental for patients and for their families, and from the societal perspective it is crucial that we come up with new strategies that will prove more effective.
Medscape: Are you and your colleagues working exclusively with heart failure patients or are you working in other specialties?
Dr. Chaudhry: I am a general internist by training, so I am very interested in this issue of hospitalizations and readmissions for patients with a number of different diseases, including myocardial infarction and community-acquired pneumonia.
Medscape: Are the data relevant from one specialty to another?
Dr. Chaudhry: There are some relevant messages. I think particularly the issue of the complexity of patients and the challenges of transitioning from hospital to home are relevant themes for patients with a number of different diagnoses, including heart failure.
