Patients
209 patients, with ages ranging from 37 to 96 years, consented to take part in this study. 84 patients agreed to undergo telemonitoring and 125 to receive usual care. All patients completed the 12-month interview. Medical-record reviews for deaths, admissions and vital-status verification were completed for 100% of patients.
The mean age of patients who opted into being part of an active monitoring cohort was 74.30 ± 12.48 years and in the control cohort it was 77.82 ± 9.23 years. This is an older population than other contemporary studies (76.43+/-10.73 in our study v 63.80+/- 11.40 in McMurray et al (5)).
Despite this they had lower NYHA scores (Figure 1).
There were no significant differences between the cohorts in terms of:
- New York Heart Association (NYHA) scores
- Ejection fraction
- Systolic blood pressure (BP)
- Age of patients
- Gender of patients
- Weight of patients
The absence of confounding variables allows us to attribute any differences observed between the two cohorts to the active monitoring protocol.
Mortality rate analysis
The overall cardiovascular mortality rate during the one year duration of the study was 5.74%. This corresponds to 12 cardiovascular deaths; 5 of which occurred in the active monitoring group and 7 occurring in the control group. This resulted in one-year mortality rates of 6.02% and 5.56% for the active and control cohorts respectively. There was no significant difference between the two groups.
We further explored the hypothesis that active monitoring may have yielded improved outcomes in terms of the individual patient’s Seattle Heart failure scores and their corresponding predicted one year survival probability. This allowed us to validate if the number of observed deaths was significantly higher or lower than predicted by the Seattle Heart failure scores. No significant differences were found, indicating that for both cohorts the number of observed cardiovascular deaths was in line with the predictions made using the Heart Failure score.
All deaths due to cardiovascular causes occurred in patients over 70 years of age.
Admission rate analysis
The difference in admission rate to hospital for heart failure between active and control cohorts was also studied. Throughout the duration of the pilot study, a total of 11 admissions occurred; 9 of which occurred within the actively monitored group and 2 in the control cohort. This corresponds to a 1-year heart failure admission rate of 10.84% and 1.59% for the active and control groups respectively (p-value of 0.0078) and 5.2% for the whole study group. These admission rates are similar to those in the PARADIGM HF study, where 14% of patients were admitted with heart failure after 27 months. Admitted patients had an average age of 77.72 ± 6.93, with all but one admitted patient being over the age of 70. The actively monitored cohort had a significantly higher rate of hospital admissions.
Drug prescription analysis
Throughout the duration of the study, patients were expected to be on evidence based drugs for the treatment of heart failure (Beta blockers, ACEI/ARB and mineralocorticoid antagonists).
As patients were taking different formulations of the various drug types we standardised the doses using dose equivalence tables (Table 1).
Table 1.
Beta-blocker
|
Dose level 1
|
Level 2
|
Level 3
|
Level 4
|
Bisoprolol
|
1.25mg
|
2.5mg
|
5mg
|
10mg
|
Carvedilol
|
3.125mg bd
|
6.25 mg bd
|
12.5mg bd
|
25mg bd
|
Metoprolol
|
12.5mg bd
|
25mg bd
|
50mg bd
|
100mg bd
|
ACE Inhibitor
|
Dose level 1
|
Level 2
|
Level 3
|
Level 4
|
Ramipril
|
1.25mg
|
2.5mg
|
5mg
|
10mg
|
Lisinopril
|
5mg
|
10 mg
|
20mg
|
40mg
|
Perindopril
|
2mg
|
4mg
|
8mg
|
16mg
|
A2 receptor blocker
|
Dose level 1
|
Level 2
|
Level 3
|
Level 4
|
Candesartan
|
4mg
|
8mg
|
16mg
|
32mg
|
Valsartan
|
20mg bd
|
40mg bd
|
80mg bd
|
160mg bd
|
Losartan
|
25mg
|
50mg
|
100mg
|
150mg
|
MRA
|
Dose level 1
|
Level 2
|
Level 3
|
Level 4
|
Spironolactone
|
25mg
|
50mg
|
100mg
|
|
Eplerenone
|
25mg
|
50mg
|
100mg
|
|
Table 2 below characterizes the changes in drug prescription which occurred during the study, for each of the four prescription drugs considered. With the exception of ACE Inhibitors, we observe a greater increase in prescription dosage for patients in the active cohort as compared to the control cohort. Furthermore, the results indicate that for all drugs considered, a larger percentage in the actively monitored group of patients concluded the study on each of the individual drugs and on all 3 classes of drugs (62% v 30%).
Table 2.
Drug
|
Dose change (level)
|
Median Final Dose
|
Prescribed at study conclusion
|
Beta-blocker active
|
0.66
|
3.38mg
Bisoprolol equivalent
|
92%
|
Beta-blocker control
|
0.48
|
3.75mg
|
52%
|
ARB Active
|
-1.22
|
12.5mg Candesartan equivalent
|
22%
|
ARB control
|
-2.3
|
14.25mg
|
19%
|
ACE-I active
|
0.13
|
5mg Ramipril equivalent
|
70%
|
ACE-I control
|
0.27
|
5mg
|
29%
|
MRA active
|
0.67
|
25mg Spironolactone equivalent
|
62%
|
MRA control
|
0.21
|
25mg
|
30%
|
At the end of the study, in the active group v control group, 92% v 52% of patients were on a beta-blocker (significant difference; p-value 4.84 x 10-10), 92% v 48% on ACE-I/ARB (significant difference; p-value 1.59 x 10-11), and 62% v 30% on an MRA (significant difference; p-value 8.46 x 10-6). There were no significant differences in the final doses achieved.
Overall satisfaction with active monitoring delivery
A questionnaire was provided to all patients who opted to receive active monitoring. The purpose of the questionnaire was to determine overall satisfaction levels with the monitoring and identify areas for potential improvement. A questionnaire was circulated at the midpoint of the study (6 months) as well as at its conclusion (12 months). Of all patients in the active control group, a total of 27 completed both questionnaires (corresponding to 32.53% of all patients under active monitoring). The characteristics of this subset of the active cohort were consistent with the overall characteristics of all patients in the study; there was no significant differences in cardiovascular status (e.g., NYHA scores) or personal characteristics (e.g., gender, age).The questionnaire consisted of a eleven questions measuring the patient satisfaction on various aspects of the active monitoring treatment. The overall response from patients was positive, with an average satisfaction of 8.54 / 10. Furthermore, when asked how likely they would be recommend the active monitoring treatment, patients gave an average score of 8.34 / 10.
- Satisfaction with verbal responses of GPs dropped from the 6 month to the 12 month questionnaire as demonstrated by Figure 2.
- There was also a significant drop in satisfaction with the website, as visualized by Figure 3. Moreover, the “Advice for everyday life” section of the website also saw a significant drop in satisfaction, indicating that this section of the website may require expanding or updating (figure 4).
- Satisfaction with the booklet “Understanding heart failure” also saw a significant drop from the 6 month to 12 month questionnaire (figure 5). This may indicate that there is a need for the booklet to provide further guidance and advice for patients during the latter stages of treatment.
Adverse Events
No adverse events were reported during the study period.