If home telemonitoring reduces mortality in heart failure,is this just due to better guideline-based treatment?

Abstract

To investigate, in a ‘real-world’ setting, the impact of home telemonitoring (HTM) compared to usual care on achieved dose of guideline-recommended medication, hospitalisation rate and mortality in patients with heart failure (HF). Methods: We retrospectively analyzed data on 333 patients with HF referred to a HTM service supported by a nurse-specialist (mean age 71±12 years, mean left ventricular ejection fraction (LVEF) 36 ± 11% and median N-Terminal pro B-type Natriuretic Peptide (NT-proBNP) 2,972 ng/L (interquartile range (IQR): 1,447–7,801 ng/L)). Most patients (n = 278) accepted HTM (HTM-group) but 55 refused and received usual care (UC-group). In the HTM-group, weight, heart rate, blood pressure and symptom severity were measured daily. Results: At referral, respectively 90%, 90%, 67% and 94% of patients with LVEF ≤40% (n = 229) were treated with β-blockers (BB), angiotensin converting enzyme-inhibitors (ACE-I) or angiotensin receptor blockers (ARB), mineralocorticoid receptor antagonists (MRA) and diuretics, with rates similar between groups. After 6 months, prescription of BB (92% vs 83%), ACE-I/ARB (92% vs 90%) and MRA (68% vs 67%) did not differ significantly between groups. The proportions of patients who achieved ≥50% and ≥100% of target doses of BB, ACE-I/ARB and MRA were also similar in each group. However, during a median follow-up of 1094 days (IQR 767–1419) fewer patients who chose HTM died (33% vs 49%; P = 0.002). Conclusion: Patients who choose HTM have a better prognosis than those who do not but this does not appear to be mediated through greater prescription of key HF medications.

Keywords

ehealth telecardiology telehealth

Introduction

Patients with heart failure (HF) pose a complex and growing problem for healthcare systems world-wide. Due to ageing and improved primary and secondary prevention of coronary heart disease, the prevalence of HF is likely to rise substantially.^1,2 Not only do patients with HF face a poor prognosis (with mortality rates similar to some common types of cancer),³ but also frequent re-hospitalisations. Some of the admissions are the result of poor discharge planning, complex medical regimens, patients’ failure to adhere to treatment and diet or failure to prescribe appropriate medical therapy at the right dose; up to two-thirds of hospitalizations may be preventable.^4–10

Several meta-analyses suggest that home telemonitoring (HTM) of symptoms and vital signs might reduce HF hospitalisations and mortality.^11–14 The mechanisms underlying these potential benefits are not well understood, but may include better self-care by increasing a patient’s knowledge and compliance with advice, improved prescription of guideline-based medication by clinicians, or earlier intervention when heart failure worsens.

The aim of the present analysis was to investigate the impact of HTM compared to usual care on prescription rates of guideline HF medication, hospitalisation rates and all-cause mortality, within the context of an established HTM service for the management of patients with HF.

Methods

Study population

Consecutive patients with HF referred between November 2007 and January 2013 to a National (UK) Health Service funded HTM service, supported by a HF nurse-specialist with access to cardiology advice, were identified. Patients were offered HTM and either agreed (HTM-group) or declined and received usual care (UC-group). Hospital records were retrospectively reviewed, and demographic, clinical and outcome data were collected.

The Kingston-upon-Hull telehealth model offers HTM to patients with a diagnosis of HF who (a) have recently been diagnosed with HF; (b) have recently been hospitalised due to HF; (c) have worsening chronic HF or (d) need frequent medication changes. Patients have to live somewhere that allows installation and operation of the HTM equipment, and be mentally and physically able to use the system properly (either alone or with the assistance of a caregiver).

Because the study population consisted, by definition, of unstable HF patients, we assumed that up-titration of medication was most likely to occur in the first months after referral. Information on drug treatment was therefore collected at baseline and after 6 months. All subjects gave written informed consent for data collection.

HTM system and its implementation

The Motiva HTM system (Philips, Amsterdam, The Netherlands) is a television-based platform that can deliver educational videos to the patient and transmit daily symptom questionnaires and measurement data (i.e. weight, heart rate and blood pressure) from the patient’s home via a broadband internet connection to a secure server. The system consists of blue-tooth connected weighing scales and an automated sphygmomanometer. Clinical alerts are generated by worsening symptoms of HF, and/or when vital signs fall outside of the individually predefined ‘normal range’.

The symptom questionnaire and measurement data can be accessed by members of the HTM team using a web-based patient management system (PMS). Technical (i.e. co-ordination of unit installs, fault reporting, dealing with missed measurements and validation of out-of-range measurements) and clinical triage (i.e. responding to validated alerts, ringing the patient, implementing protocol driven diuretic changes and escalating clinical alerts) are both provided by an HTM nurse.¹⁵ Clinical alerts are dealt with by the HTM nurse calling the patient and then, if necessary, a clinical responder; either a community HF nurse with prescribing qualifications, a GP or a cardiologist if long-term changes in therapy are required.

Patients in the HTM group were asked to make daily measurements. Failure or reluctance to use the system (e.g. because of technical problems or malcompliance) generated an alert, and the patient was then contacted by the HTM nurse. If the patient was stable at 4 months, HTM system removal was offered.

Statistical analysis

Patient characteristics are expressed as means ± standard deviations (SD) for normally distributed variables or medians with interquartile ranges (IQR) for skewed data. Medication doses were quantified as percentages of target dose (as defined by the 2012 ESC guidelines on acute and chronic heart failure¹⁶). Follow-up started on the day of referral to the HTM service and was censored on 31/12/2013. Differences between groups were compared using the independent Student’s t-test or Mann-Whitney U-test (for variables not normally distributed) for continuous variables, and the Chi square test for categorical variables.

The primary outcome of interest was all-cause mortality. Secondary endpoints included HF hospitalisation rates, time-to-first HF admission and time-to-first event of the composite endpoint of all-cause mortality or HF hospitalisation. The number of HF hospitalisations was assessed one year after enrollment and compared between groups using the independent Student’s t-test. Cumulative survival was analyzed by means of the Kaplan-Meier method and compared between groups using the log-rank test. The same method was applied for time-to-first HF hospitalisation and time-to-first-event. Univariable and multivariable Cox regression analysis was used to explore the relationships between variables of interest and outcome. Variables associated with the primary outcome in univariable analysis (P < 0.1) were entered in a multivariable model and included: HTM, age, heart rate, body mass index (BMI), log-transformed N-Terminal pro B-type Natriuretic Peptide (NT-proBNP), estimated glomerular filtration rate (eGFR), haemoglobin, sodium, cardiac resynchronization therapy (CRT), angiotensin converting enzyme inhibitors (ACE-I) and/or angiotensin receptor blockers (ACE-I/ARB) at baseline and New York Heart Association (NYHA) class.

A secondary endpoint was the prescription rate of guideline-recommended HF medication. However, since beta blockers (BB), ACE-I and/or ARBs and mineralocorticoid receptor antagonists (MRA) are known to improve prognosis of patients with heart failure and a reduced LVEF (HFrEF), but not of those with heart failure and a normal ejection fraction (HFnEF), this particular analysis was limited to patients with HFrEF (LVEF ≤ 40%; N = 229). Prescription rates were assessed at baseline and after 6 months. Statistical analysis was performed using SPSS software (18.0, SPSS Inc. Chicago IL, USA). A two-tailed P-value < 0.05 was considered statistically significant.

Results

Patient characteristics

Of 333 patients referred for HTM, mean age was 71 ± 12 years, 73% were men and 278 (83%) agreed to use HTM (HTM-group) and 55 refused (UC-group). Patient characteristics are shown in Table 1. In most respects, patients who chose or refused HTM were similar but those that refused were significantly older, had more renal impairment, lower BMI and a higher heart rate.

Table 1.

–Baseline characteristics of patients. Values are expressed as percentages, mean ± standard deviation (SD) for normally distributed variables, or median with interquartile range (IQR) for skewed data. P-values represent differences between groups.

Variables	Overall (n = 333)	HTM (n = 278)	UC (n = 55)	P-value	Missing data
Demographics
Male	243 (73%)	202 (73%)	41 (75%)	0.77	0
Age (years)	71 ± 12	70 ± 11	75 ± 10	0.009	0
IHD	205 (63%)	173 (64%)	32 (58%)	0.44	6
AF	136 (41%)	115 (42%)	21 (38%)	0.63	2
Diabetes	118 (35%)	105 (38%)	13 (24%)	0.04	0
COPD/asthma	100 (30%)	85 (31%)	15 (27%)	0.62	0
Hypertension	153 (46%)	130 (47%)	23 (42%)	0.50	0
BMI (m/kg²)	27.8 (IQR 24.1–31.3)	28.1 (IQR 24.3–32.6)	26.3 (IQR 23.0–28.9)	0.001	24
CRT	40 (12%)	32 (12%)	8 (15%)	0.52	0
ICD	35 (11%)	30 (11%)	5 (9%)	0.70	0
Renal failure (eGFR<60 ml/min)	179 (54%)	143 (52%)	36 (67%)	0.05	3
Clinical data
SBP (mmHg)	116 ± 21	116 ± 21	116 ± 18	0.89	12
Heart rate (bpm)	76 ± 15	75 ± 15	81 ± 17	0.01	13
NYHA class				0.78	9
I/II	183 (57%)	154 (57%)	29 (55%)
III/IV	141 (44%)	117 (43%)	24 (45%)
Blood results
Sodium (mmol/L)	136 ± 4	136 ± 4	135 ± 4	0.45	0
Hemoglobin (g/dL)	12.6 ± 2.0	12.6 ± 2.0	12.1 ± 1.9	0.10	3
NTproBNP (ng/L)	2972 (IQR 1447–7801)	2853 (IQR 1421–7274)	4269 (IQR 1716–10455)	0.23	51
eGFR (ml/min)	57 (IQR 38–77)	59 (IQR 39–81)	48 (IQR 33–67)	0.01	3
Echocardiography
LVEF (%)	36 ± 11	36 ± 10	35 ± 11	0.41	5
LVEF ≤ 40%	229 (70%)	191 (70%)	38 (70%)	0.92	5
Medication
BB	281 (84%)	237 (85%)	44 (80%)	0.32	0
ACE-I/ARB	284 (85%)	239 (86%)	45 (82%)	0.42	0
MRA	208 (63%)	175 (63%)	33 (60%)	0.68	0
Diuretic	313 (94%)	262 (94%)	51 (93%)	0.66	0

HTM: home telemonitoring; UC: usual care; IHD: ischaemic heart disease; AF: atrial fibrillation; COPD: chronic obstructive airway disease; BMI: body mass index; CRT: cardiac resynchronization therapy; ICD: internal cardioverter defibrillator; eGFR: estimated glomerular filtration rate; SBP: systolic blood pressure; NYHA class: New York Heart Association class; NTproBNP: N-Terminal pro B-type Natriuretic Peptide; LVEF: left ventricular ejection fraction; BB: beta-blockers; ACE: angiotensin converting enzyme inhibitor; ARB: angiotensin receptor blocker; MRA: mineralocorticoid receptor antagonist.

Six patients requested the system to be removed within the first 4 months. Altogether, the HTM system was removed in 54 patients before the censoring date. The median duration of telemonitoring (in survivors) was 1027 days (IQR 624–1415).

Most patients (258; 77%) were referred for HTM following a hospitalization for HF and in most cases (205; 62%) HF had been diagnosed for the first time during this hospitalisation. Outpatient clinic visits (28; 8%) and follow-up by HF nurse-specialists (47; 14%) accounted for the remaining 75 referrals.

A total of 229 patients (191 in the HTM-group and 38 in the UC-group) had HFrEF. Differences in medical therapy were further explored in this subgroup.

Differences in medical therapy for patients with HFrEF (n = 229)

At referral, 94% of patients with HFrEF were taking loop diuretics, and, respectively 90%, 90%, 67% were treated with (any dose of) BB, ACE-I/ARB and MRA, with rates similar in each group (Table 2). However, more patients in the HTM-group received at least 50% of the target BB dose and 100% of the target MRA. After 6 months, overall prescription rates were nearly identical. The proportions of patients achieving ≥50% and ≥100% of target doses of BB, ACE-I/ARB and MRA were similar in each group (Table 2). Patients in the HTM-group were no more likely to receive treatment changes than the UC-group (Figure 1). Patients who did not reach the target dose of BB were older (Table 3). For ACE-I/ARB, failure to achieve the target dose was associated with systolic blood pressure <100 mmHg, eGFR <30 ml/min, older age and higher plasma concentrations of NTproBNP. Patients who received less than 100% of the target dose of MRA were more likely to have a low eGFR (Table 3).

Table 2.

– Proportions of patients with heart failure and reduced ejection fraction (HFrEF; n = 229) who were treated with any dose, ≥ 50% and ≥100% of target dose of beta-blockers (BB), angiotensin converting enzyme inhibitors (ACE-I) and/or angiotensin receptor blockers (ARB), mineralocorticoid receptor antagonists (MRA) and loop diuretics. Target doses were defined according to the 2012 ESC guidelines on acute and chronic heart failure.

Patients with HFrEF (LVEF ≤ 40%)
Drug	Baseline			6 months
Drug	All	HTM (n = 191)	UC (n = 38)	All	HTM (n = 174)	UC (n = 30)
BB any dose	205 (90%)	173 (91%)	32 (84%)	186 (91%)	161 (92%)	25 (83%)
BB ≥50%	75 (33%)	68 *(36%)	7 *(18%)	84 (41%)	75 (43%)	9 (30%)
BB ≥100%	32 (14%)	29 (15%)	3 (8%)	32 (16%)	29 (17%)	3 (10%)
ACE/ARB any dose	207 (90%)	173 (91%)	34 (90%)	186 (92%)	159 (92%)	27 (90%)
ACE/ARB ≥50%	128 (56%)	106 (56%)	22 (58%)	108 (53%)	92 (53%)	16 (53%)
ACE/ARB ≥100%	60 (26%)	49 (26%)	11 (29%)	61 (30%)	54 (31%)	7 (23%)
MRA any dose	153 (67%)	127 (67%)	26 (68%)	138 (68%)	118 (68%)	20 (67%)
MRA ≥50%	150 (66%)	124 (65%)	26 (68%)	133 (66%)	113 (65%)	20 (67%)
MRA ≥100%	70 (31%)	64 *(34%)	6 *(16%)	63 (31%)	57 (33%)	6 (20%)
Diuretic any dose	215 (94%)	179 (94%)	36 (95%)	190 (94%)	161 (93%)	29 (97%)

P < 0.05.

HTM: home telemonitoring; UC: usual care; LVEF: left ventricular ejection fraction.

Table 3.

Differences between groups of patients with heart failure and reduced ejection fraction (HFrEF;n = 229) receiving < or ≥ 100% of target doses of neurohormonal agents after 6 months of follow-up.

Target dose ≥100% at 6 months for patients with HFrEF
	Yes	No	P-value
BB	32 (16%)	173 (84%)
> = 50%	NA	52 (30%)
SBP < 100 mmHg	7 (23%)	34 (20%)	0.74
Heart rate < 65 bpm	9 (29%)	30 (18%)	0.14
Heart rate >70bpm	17 (55%)	99 (58%)	0.73
COPD/asthma	8 (25%)	51 (30%)	0.61
Age (years)	65 ± 15	70 ± 12	0.02
NTproBNP (ng/L; median and IQR)	2340 (1378–6956)	3225 (1573–8216)	0.53

ACE-I/ARB	61 (30%)	142 (70%)
> = 50%	NA	47 (33%)
SBP < 100 mmHg	5 (8%)	36 (26%)	0.005
Potassium ≥ 5 mmol/L	8 (13%)	15 (11%)	0.60
eGFR < 30 ml/min	2 (3%)	25 (18%)	0.006
Age (years)	66 ± 12	71 ± 12	0.02
NTproBNP (ng/L; median and IQR)	2241 (1151–4799)	3844 (1810–9662)	0.01

MRA	63 (31%)	140 (69%)
> = 50%	NA	70 (50%)
SBP < 100 mmHg	17 (27%)	24 (18%)	0.13
Potassium ≥ 5 mmol/L	7 (11%)	16 (11%)	0.95
eGFR < 30ml/min	4 (6%)	23 (16%)	0.05
Age (years)	69 ± 11	70 ± 13	0.56
NTproBNP (ng/L; median and IQR)	3844 (1329–9590)	2884 (1612–7123)	0.73

BB: beta-blockers; SBP: systolic blood pressure; COPD: chronic obstructive airway disease; NTproBNP: N-Terminal pro B-type Natriuretic Peptide; IQR: interquartile range; ACE-I: angiotensin converting enzyme inhibitor; ARB: angiotensin receptor blocker; MRA: mineralocorticoid receptor antagonist; eGFR: estimated glomerular filtration rate.

Survival

The median duration of follow-up of survivors was 1094 days (IQR 766–1419). Cardiovascular and all-cause mortality at 12 months were 13% and 17% amongst patients who chose HTM, compared to 16% and 27% in those that refused (P 0.55 and 0.07). The Kaplan Meier curves for all-cause and cardiovascular mortality are shown in Figures 2 and 3. A total of 36 deaths were due to non-cardiovascular causes: 2 renal failure, 1 bowel perforation, 9 cancer, 2 respiratory failure (COPD), 2 falls/fracture, 17 infection/sepsis, 3 unknown.

On univariable Cox-regression analysis, patients who chose HTM had lower all-cause mortality compared to patients in the UC-group (HR (hazard ratio) 0.51, 95% CI (confidence interval) 0.33-0.79; P = 0.003). In the multivariable Cox model, choosing HTM was independently associated with improved survival (HR 0.41, 95% CI 0.25–0.68; P < 0.001). Higher NTproBNP values, presence of a CRT device and higher NYHA class were also independent predictors of mortality (Table 4). The worse outcome associated with CRT is probably due to selection bias: those patients with CRT had worse heart failure. Because the effect of HTM on titration of guideline-recommended medication was explored in patients with HFrEF only, we repeated survival analysis for this subgroup (Figure 2B). Univariate and multivariate Cox regression analysis confirmed the association between HTM and improved survival.

Table 4.

Univariable and multivariable Cox regression models for endpoint of all-cause mortality. For continuous variables, the hazard ratio is that associated with a unitary increase in that variable; for categorical variables the hazard ratio is that of ‘yes’ versus ‘no’, unless otherwise specified.

Variables	Univariable analysis			Multivariable analysis
Variables	HR (95% CI)	Wald χ²	p-value	HR (95% CI)	Wald χ²	p-value
HTM	0.51 (0.33–0.79)	9.11	0.003	0.41 (0.25–0.68)	12.31	<0.001
Sex (female vs male)	0.89 (0.59–1.34)	0.29	0.59
IHD	1.28 (0.88–1.90)	1.65	0.20
Age (year)	1.05 (1.03–1.07)	23.84	<0.001	1.03 (1.00–1.06)	3.58	0.06
NYHA (III/IV vs I/II)	1.51 (1.05–2.17)	4.91	0.03	1.92 (1.24–2.96)	8.64	0.003
SBP (mmHg)	0.99 (0.98–1.01)	0.82	0.37
Heart rate (bpm)	0.99 (0.97–1.00)	4.84	0.03	0.99 (0.97–1.00)	2.77	0.10
AF	0.88 (0.61–1.27)	0.47	0.50
Diabetes	0.77 (0.52–1.12)	1.86	1.17
BMI (kg/m²)	0.97 (0.93–0.99)	4.22	0.04	1.01 (0.97–1.06)	0.32	0.57
Log NTproBNP	2.63 (1.79–3.91)	22.77	<0.001	2.12 (1.31–3.42)	9.44	0.002
eGFR (ml/min)	0.98 (0.98–0.99)	20.56	<0.001	0.98 (0.87–1.09)	0.17	0.68
Hemoglobin (g/dL)	0.87 (0.79–0.95)	9.15	0.002	0.94 (0.82–1.07)	0.93	0.34
Sodium (mmol/L)	0.96 (0.92–0.99)	5.20	0.02	0.97 (0.93–1.01)	1.82	0.18
CRT	2.00 (1.26–3.18)	8.55	0.003	1.98 (1.14–3.44)	5.80	0.02
ICD	1.15 (0.65–2.05)	0.23	0.63
LVEF (%)	1.07 (0.91–1.25)	0.7	0.40
ACE-I/ARB (baseline)	0.52 (0.33–0.81)	8.26	0.004	0.76 (0.45–1.29)	1.04	0.31
BB (baseline)	0.79 (0.50–1.24)	1.05	0.31
MRA (baseline)	1.17 (0.81–1.70)	0.69	0.41

HR: Hazard ratio; CI: confidence interval; HTM: home telemonitoring; IHD: ischemic heart disease; NYHA: New York Heart Association; SBP: systolic blood pressure; AF: atrial fibrillation; BMI: body mass index; logNTproBNP: logarithmic transformation of N-Terminal pro Brain Natriuretic Peptide; eGFR: estimated glomerular filtration rate; CRT: cardiac resynchronization therapy; ICD: internal cardioverter defibrillator; LVEF: left ventricular ejection fraction; ACE-I/ARB: angiotensin converting enzyme-inhibitor/angiotensin receptor blocker; BB: beta-blocker; MRA: mineralocorticoid receptor antagonist.

The number of HF admissions per patient was similar in each group after 1 year (0.3 ± 0.6 for the HTM-group vs 0.2 ± 0.4 for the UC-group; P 0.51). In the HTM-group, 51 patients (18%) experienced a total of 72 HF hospitalizations within 1 year after enrollment, compared to 11 HF hospitalizations in 10 patients (18%) in the UC-group. There was no difference in time-to-first HF hospitalisation (P = 0.37) or time-to-first-event of the composite endpoint of all-cause mortality or HF hospitalisation (P = 0.38).

Discussion

In this study, HTM was associated with improved survival. There was a trend towards a significant difference in cardiovascular mortality, again in favor of the HTM group, but with a smaller number of events. Patients in the HTM group were younger and had higher BMI and eGFR, both associated with a better prognosis in patients with heart failure.¹⁷ However, HTM remained an independent predictor of improved survival even after adjustment for these possible confounding variables. Similar to previous studies that have shown a benefit from HTM, the patient population in this analysis consisted of high risk patients, most of them recently discharged from hospital after an admission for HF.^18–21

BB, ACE-I/ARB and MRA are recommended routinely for patients with HFrEF to reduce hospitalisations and mortality.¹⁶ Some studies suggest that some of this benefit is dose-dependent.^22–25 However, despite current recommendations, these life-prolonging agents remain generally underused, both in prescription rates and in target doses achieved.^26,27 With HTM, physicians and nurses can receive daily information on patients’ symptoms and vital signs. Theoretically, this provides an opportunity to optimize HF treatment safely and efficiently. It might thus be expected that patients on HTM are more likely to receive (target doses of) guideline-recommended medication compared to usual care. Moreover, it could be hypothesized that better medical treatment is one of the mechanisms underlying the benefit of HTM.

Most patients in our study had HFrEF, which is probably due to the fact that heart failure with preserved ejection fraction is often not recognized as such, especially when patients are admitted to non-cardiology wards. For patients with HFrEF, we found a similar and remarkably high rate of prescription of guideline-recommended HF medication amongst patients who did or did not chose HTM. The proportion of patients already treated with ACE-I/ARB, BB and MRA at baseline was higher compared to previously reported data.^26–28 However, the number of patients achieving ≥50% and ≥100% of guideline-recommended target doses was low during follow-up, especially for BB. The mean heart rate at 6 months in patients with HFrEF and sinus rhythm was 68 ± 13 bpm in patients taking ≥ 100% of target BB dose, compared to 72 ± 13 bpm in those with lower doses (P 0.18; supplementary table), suggesting that, based on heart rate, there was opportunity for further BB up-titration. Although the HF nurse-specialists supporting the HTM service have access to specialist medical advice, they themselves are only allowed to provide advice on diuretics and even then only short-term changes. Routine medical support (e.g. through weekly case discussions, or even daily) and/or implementation of decision-support systems might enhance up-titration of HF medication.

While other trials have shown that HTM is associated with higher prescription rates of disease-modifying therapies,^19,29 this does not seem to be the explanation for the better prognosis with HTM in HFrEF patients included our study. It may be that HTM led to better compliance with prescribed medications or lifestyle advice and patient empowerment, and/or earlier intervention (e.g. referral to the GP or cardiologist).

Although BB reduce morbidity and mortality in patients with HFrEF, their benefits may be confined to patients in sinus rhythm.³⁰ This might also be true for CRT.³¹ More than 40% of our patients were in AF, a large subgroup of patients with HFrEF who may obtain little or no benefit from key interventions for HF.

Three interventions that are known to improve the prognosis of patients with HFrEF were associated with trends to a worse outcome in this study; CRT, ICD and MRA. This may be because sicker patients are more likely to receive these interventions again demonstrating that the outcome with treatment and the response to it are not the same.^32,33 Also, patients who responded well to these treatments may be much less likely to be referred for HTM. It is perhaps not unexpected that patients who fail to respond well to these interventions are going to have a bad prognosis. Great caution is required in interpreting observational trials of any intervention, including HTM.

In this study, most patients accepted HTM when offered (83%). Although the possibility of having the equipment removed was pro-actively discussed after 4 months of follow-up, most patients preferred to continue HTM. After 1 year, 90% of 230 patients alive in the HTM-group were still using the system with high rates (>80%) of daily measurement compliance. Similarly high rates of usage were previously reported in the TEN-HMS trial, which investigated the same technology.¹⁹

Study limitations

Our study is limited by the retrospective, observational design with possibility of selection bias. Randomized trials - although providing more robust evidence – also have limitations, including strict inclusion/selection criteria, selection of better educated patients with fewer comorbidities and lack of integration into existing services. Our aim was to assess the effects of HTM in an established service setting. Propensity matching might have reduced the likelihood of bias,³⁴ but the sample size was too small to perform such an analysis. Although observational data cannot establish a cause-effect relationship, we did find a strong relation between the use of HTM and improved survival, whatever the underlying mechanism may be.

The study did not have an independent end-points committee to determine cause of death. For this reason, and because of the smaller number of events, we believe that all-cause mortality is a better indicator of benefit of the intervention than cardiovascular mortality.

In those patients with HFrEF who died before 6 months (17 in the HTM-group and 8 in the UC-group) information on achieved doses of guideline-recommended medication was not available.

When given the choice, a surprisingly low proportion of patients wanted to discontinue HTM. It would have been interesting to identify the reasons for discontinuation, but unfortunately the reasons were not systematically documented. This is likely to be a mixed group of patients who either have advanced HF and no longer feel able to cope or have improved and no longer feel they need it or have simply become bored with the routine of repeat measurements.

Conclusions

Patients who accept HTM when offered the choice have a lower mortality. This was not explained by prescription of better pharmacological treatment but might reflect better adherence to prescribed therapy. Differences may have occurred by chance or reflect differences in the psychology of patients who do or don’t accept medical recommendations or to changes in patient behaviour caused by HTM. In routine clinical practice, HTM often may not be sufficient by itself to achieve target doses recommended by guidelines. Service audits, facilitated by decision-support applications, may improve effective delivery of care.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Stewart

. Heart failure and the aging population: an increasing burden in the 21st century? Heart 2003; 89(1): 49–53.

Heidenreich

. Forecasting the impact of heart failure in the United States: a policy statement from the American Heart Association. Circ Heart Fail 2013; 6(3): 606–19.

Stewart

. More ‘malignant' than cancer? Five-year survival following a first admission for heart failure. Eur J Heart Fail 2001; 3(3): 315–22.

Greenberg

. Acute decompensated heart failure - treatments and challenges. Circ J 2012; 76(3): 532–43.

Ashton

. The association between the quality of inpatient care and early readmission. Ann Intern Med 1995; 122(6): 415–21.

Haldeman

. Hospitalization of patients with heart failure: National Hospital Discharge Survey, 1985 to 1995. Am Heart J 1999; 137(2): 352–60.

Youn

. Treatment performance measures affect clinical outcomes in patients with acute systolic heart failure: report from the Korean Heart Failure Registry. Circ J 2012; 76(5): 1151–8.

Hernandez

. Relationship between early physician follow-up and 30-day readmission among Medicare beneficiaries hospitalized for heart failure. Jama 2010; 303(17): 1716–22.

Moser

Mann

. Improving outcomes in heart failure: It's not unusual beyond usual care. Circulation 2002; 105(24): 2810–2.

10.

Michalsen

König

Thimme

. Preventable causative factors leading to hospital admission with decompensated heart failure. Heart 1998; 80(5): 437–441.

11.

Polisena

. Home telemonitoring for congestive heart failure: a systematic review and meta-analysis. J Telemed Telecare 2010; 16(2): 68–76.

12.

Clarke

Shah

Sharma

. Systematic review of studies on telemonitoring of patients with congestive heart failure: a meta-analysis. J Telemed Telecare 2011; 17(1): 7–14.

13.

Inglis

. Which components of heart failure programmes are effective? A systematic review and meta-analysis of the outcomes of structured telephone support or telemonitoring as the primary component of chronic heart failure management in 8323 patients: Abridged Cochrane Review. Eur J Heart Fail 2011; 13(9): 1028–40.

14.

Xiang

Liu

. Meta-analysis and meta-regression of telehealth programmes for patients with chronic heart failure. J Telemed Telecare 2013; 19(5): 249–59.

15.

Crundall-Goode

Goode

. Using Telehealth for Heart Failure: Barriers, Pitfalls and Current Nursing Service Models. British Journal of Cardiovascular Nursing 2014; 9(18): 287–297.

16.

McMurray

. ESC guidelines for the diagnosis and treatment of acute and chronic heart failure 2012: The Task Force for the Diagnosis and Treatment of Acute and Chronic Heart Failure 2012 of the European Society of Cardiology. Developed in collaboration with the Heart Failure Association (HFA) of the ESC. Eur J Heart Fail 2012; 14(8): 803–69.

17.

Lavie

. Impact of Obesity and the Obesity Paradox on Prevalence and Prognosis in Heart Failure. JACC Heart Fail 2013; 1(2): 93–102.

18.

Goldberg

. Randomized trial of a daily electronic home monitoring system in patients with advanced heart failure: the Weight Monitoring in Heart Failure (WHARF) trial. Am Heart J 2003; 146(4): 705–12.

19.

Cleland

. Noninvasive home telemonitoring for patients with heart failure at high risk of recurrent admission and death: the Trans-European Network-Home-Care Management System (TEN-HMS) study. J Am Coll Cardiol 2005; 45(10): 1654–64.

20.

Abraham

. Wireless pulmonary artery haemodynamic monitoring in chronic heart failure: a randomised controlled trial. Lancet 2011; 377(9766): 658–66.

21.

Dendale

. Effect of a telemonitoring-facilitated collaboration between general practitioner and heart failure clinic on mortality and rehospitalization rates in severe heart failure: the TEMA-HF 1 (TElemonitoring in the MAnagement of Heart Failure) study. Eur J Heart Fail 2012; 14(3): 333–40.

22.

Packer

. Comparative effects of low and high doses of the angiotensin-converting enzyme inhibitor, lisinopril, on morbidity and mortality in chronic heart failure. ATLAS Study GrouCirculation 1999; 100(23): 2312–8.

23.

Konstam

. Effects of high-dose versus low-dose losartan on clinical outcomes in patients with heart failure (HEAAL study): a randomised, double-blind trial. Lancet 2009; 374(9704): 1840–8.

24.

Verbrugge

. Uptitration of renin-angiotensin system blocker and beta-blocker therapy in patients hospitalized for heart failure with reduced versus preserved left ventricular ejection fractions. Am J Cardiol 2013; 112(12): 1913–20.

25.

Fiuzat

. Association between adrenergic receptor genotypes and beta-blocker dose in heart failure patients: analysis from the HF-ACTION DNA substudy. Eur J Heart Fail 2013; 15(3): 258–66.

26.

Komajda

. The EuroHeart Failure Survey programme–a survey on the quality of care among patients with heart failure in Europe. Part 2: treatment. Eur Heart J 2003; 24(5): 464–74.

27.

de Groote

. Is the gap between guidelines and clinical practice in heart failure treatment being filled? Insights from the IMPACT RECO survey. Eur J Heart Fail 2007; 9(12): 1205–11.

28.

Fonarow

. Influence of a performance-improvement initiative on quality of care for patients hospitalized with heart failure: results of the Organized Program to Initiate Lifesaving Treatment in Hospitalized Patients With Heart Failure (OPTIMIZE-HF). Arch Intern Med 2007; 167(14): 1493–502.

29.

Antonicelli

. Impact of home patient telemonitoring on use of beta-blockers in congestive heart failure. Drugs Aging 2010; 27(10): 801–5.

30.

Rienstra

. Beta-Blockers and Outcome in Heart Failure and Atrial FibrillationA Meta-Analysis. JACC: Heart Failure 2013; 1(1): 21–28.

31.

Cleland

. An individual patient meta-analysis of five randomized trials assessing the effects of cardiac resynchronization therapy on morbidity and mortality in patients with symptomatic heart failure. Eur Heart J 2013; 34(46): 3547–56.

32.

Cleland

Ghio

. The determinants of clinical outcome and clinical response to CRT are not the same. Heart Fail Rev 2012; 17(6): 755–66.

33.

Cleland

. Should we be trying to define responders to cardiac resynchronization therapy? JACC Cardiovasc Imaging 2010; 3(5): 541–9.

34.

Goode

Crundall-Goode

Barrett

. Measuring the effectiveness of home telemonitoring. British Journal of Healthcare Management 2014; 20(4): 165–171.