Result card

  • ORG8: What are the likely budget impacts of implementing the technologies being compared?
English

What are the likely budget impacts of implementing the technologies being compared?

Authors: Valentina Prevolnik Rupel, Taja Čokl, Eleftheria Karampli

Internal reviewers: Ulla Saalasti - Koskinen, Elle Kisk, Ricardo Ramos

To answer the questions in the assessment elements we mainly used the basic literature search provided for the whole project. Additionally, two more systematic searches were used: one performed by ORG and ECO domains (described in methodology of ECO domain) and one perfomed by EFF, SAF and ECO domains (described in SAF domain).The results are provided in descriptive way.

While some studies reported {1} {2}{23} no statistically significant difference in healthcare costs (either total costs or all-cause hospital costs), other studies reported important and significant reductions in costs. While the average costs of intervention across the studies amounted from $23,6 to $443, the reported savings amounted from $30,9 to $536 per patient per month. The savings across studies were reported in various ways which makes them hardly comparable (percentage reduction in inpatient costs, percentage reduction in overall costs, percentage reduction in total health expenditures, reduction in different currencies per patient, per nurse, per year, per month, per 6 months...). However, more important than this is the method of costs calculation that varies widely across the studies. In most studies, only direct costs are included, mostly connected to reduction in hospitalizations. More than 70 % of the studies did not take into account expenses in one of the following categories:  healthcare sector, other sectors, patient/family expenses or productivity losses. None of the studies analyzed a shift of cost, from specialits to HF nurse to GP, for instance. In 80 % of the studies the source and methods of the evaluations were not clear. Authors mostly focused on direct costs while omitting indirect (i.e., productivity gains and losses) and intangible costs (i.e., relief from pain, lost leisure time for patient or family) {310}. Principally, the costs were missing across majority of the studies and those of the intervention overheads, training of personnel, and patient related costs.  There is a difficulty in capturing all of the effects of telehealth intervention. Thus the cost effectiveness evidence for specific implementations in the field of telehealth is limited. Problems with telehealth interventions reside in absence of quality data and appropriate measures. The quality of economic data is especially questionnable. The quality of evidence in the scientific literature is poor. More studies on all costs are needed to reach the unbiased conclusion.

1 of the 16 studies on STS reviewed by Inglis et al. {130} reported the effect of the intervention on the cost of care. Two studies {2}{23} found no statistically significant difference in healthcare costs (either total costs or all-cause hospital costs) in a timeframe of 6-18 months. Three studies {24} {25} {26} reported reductions in cost (either cost per admission or overall reduction in healthcare costs). Results from RCTs where remote monitoring interventions were offered alongside existing services potentially underestimate the interventions' effectiveness and cost-effectiveness in comparison to integrating these services into existing ones {130}. In Wakefield et al {1} the differences in resource use (hospital stays, hospital days, emergency department visits) were not significant. Clark et al {1280} included 14 RCTs on STS or TM in a review. STS included monitoring of symptoms, medicine management, and education and counselling on lifestyle. In 4 studies the effect on cost per patient:  in Riegel et al. {26} a 46% reduction in inpatient costs (p<0,04) was reported. The costs of intervention amounted to $443 per patient. In Laramee et al {24} $2.482 average reduction per patient was noticed – the costs amounted to $228,52 per patient. In Tsuyuki et al.  {28} $2.531 cost reduction per patient was noticed. In Riegel et al. {23} no effect on cost of care were detected.  The average costs of intervention in Barth et al. {28} amounted to $23,60 per patient.

Although HF DM improves the quality of care and decreases hospitalization for patients with HF (the primary driver of cost for HF care) in a number of studies, the impact on cost is less certain. Although many of the RCT of HM DM provided personnel costs, few included cost estimates for patients medications (increased utilization of medications), patient time, additional clinic visits potentially generated by the intervention, patient materials, and personnel training.  Study of telephone follow up {1500} estimated that training involved 95 hours of personnel time at $ 26,51 per hour, training cost $ 2.518 per case manager. In addition, the increased surveillance of patients by interventions identifies a number of issues that may have gone undetected and subsequently increase physician time. One economic model of HF DM based in the United Kingdom, wich attempted to account for the cost of the interventions, changes in pharmacotherapy, and increases in clinic visits, estimated that 49.000 British pounds per year could be generated for each HF nurse hired to provide a HF DM intervention {1500}.

 

The study by Chen et al {13} was designed to assess the clinical effect of a home-based telephone intervention in Chinese heart failure patients. A total of 550 Chinese heart failure patients were enrolled into either (i) a group that received the usual standard of care (UC group); or (ii) a group that received a home-based heart failure centre management programme using nursing specialist-led telephone consultations (HFC group). The impact of the home-based intervention on medical costs over 6 months was measured {13}.

 

Table 2: Financial impact of the different types of follow-up care offered to patients with heart failure: usual

standard follow-up care (UC group) or the home-based Heart Failure Centre management

programme using nursing specialist-led telephone consultations (HFC group)        

                         

Financial measure

UC group

n = 275

(USD per patient)

HFC group

n = 275

(USD per patient)

 

Change

(%)

 

Statistical significance

Out-patient cost, 6 months

321 + 354

510 + 424

+58,9 %

P < 0,001

Total cost of all-cause in-patient heart failure care, 6 months

8.280 + 14.446

5.479 + 12.547

   -33,8 %

P = 0,02

Cost of in-patient heart failure care, 6 months

5.332 + 13.276

3.200 + 9.815

-40 %

P = 0,03

Cost of in-patient non-heart failure care, 6 months

2.948 + 7.588

2.279 + 8.446

-22,7 %

NS

Emergency department cost,

6 months

121 + 255

51 + 164

-58,1 %

P < 0,001

Total overall cost,

6 months

8.722 + 14.385

6.040 + 12.500

-30,8 %

P = 0,02

Total overall cost/month

1.454 + 2.397

1.006 + 2.083

 

 

Data show mean + SD.

NC not statistically significiant (P > 0.005).

 

The overall costs per patient for all-cause inpatient care and for in-patient care due to heart failure were both significantly lower for the HFC group compared with the UC group (P = 0,02 and P = 0,03, respectively) (Table 2). In contrast, the overall out-patient cost during the 6 months of follow-up was significantly higher per HFC patient compared with the cost per UC patient (P < 0,001). When considering all of the costs, despite having 58,9% higher out-patient care costs, the HFC home-based intervention still reduced the overall healthcare expenditure by 30,8% compared with the usual care programme.

 

Table 3: Univariate and multivariate analyses of the impact of disease management with the home-based

Heart Failure Centre management programme using nursing specialist-led telephone

consultations on clinical outcomes and healthcare costs in Chinese patients with heart failure      

 

Out-patient cost

 

Cost of all-cause

in-patient care

Cost of heart failure in-patient care

Cost of in-patient non-heart failure care

Total

overall cost

 

                Statistical

 β             significance

                 Statistical

     β        significance

                 Statistical

    β        significance

             Statistical

   β       significance

         Statistical

   β      significance

Model 1

0,236        P < 0,001

-0,103    P = 0,016

-0,091   P = 0,033

   -0,042        NS

-0,099    P = 0,020

Model 2

0,273        P < 0,001

-0,170    P = 0,001

-0,143   P = 0,004

   -0,080        NS

-0,166    P = 0,001

Model 3

0,226        P < 0,001

-0,102    P = 0,043

  -0,092      NS

   -0,040        NS

-0,100    P = 0,048

NS, not statistically significant (P > 0,05)

a Model 1, univariate analysis; model 2, multivariate analysis after adjusting for age, gender, left ventricular ejection fraction, coronary artery disease, hypertension, diabetes mellitus, hyperlipidaemia, and smoking; model 3, multivariate analysis as in model 2 but also adjusting for medication use (spironalactone, β-blockers, angiotensin-converting enzyme inhibitors or angiotensin receptor blockers, diuretics, statins and amiodarone).

 

The present study demonstrated that a home-based intervention led by nursing specialists resulted in a significantly lower admission rate, shorter hospital stay and lower healthcare expenditure. This study demonstrated a mean cost saving of $ 448 per patient per month during the 6-month DM programme. The RM reduced the costs of the treatment of inpatients by reducing the duration and risk of hospitalization. However, they did not analyze the impact of patient’s adherence on the cost of telemedicine {13}{250}.

In terms of the economic benefits, a meta analysis by Philips et al.  {13} showed that a disease management programme saved $ 359 per patient per month in a non-USA trial, while this type of intervention saved $ 536 per patient per month in a study in the USA {13}.

A study {1000} was set up to estimate the cost-effectiveness of home TM or STS strategies versus UC for adults recently discharged (within 28 days) after a HF exacerbation in England and Wales. A Markov model was used to evaluate a) STS via human to machine (HM) interface, b) STS via human to human (HH) contact, and c) TM, against d) UC. The results for base case monthly costs per patient were: GBP 27 for UC, GBP 119 for STS HM, GBP 179 for STS HH and GBP 175 for TM {1000}.

In a study by DeBusk and colleagues {15}, UC group of HF patients was compared to intervantion group who received physician-directed, nurse-managed home based program for HF, which included initial educational session, including a videotape, baseline telephone counselling session, nurse-initiated follow up telephone contacts, pharmacologic management and nurse initiated communication with physician. Nurse care managers spent an average of 9 hours per patient during the first year. No statistically significant difference between UC alone and UC supplemented with nurse management was found in the rate of rehospitalization or in the combined outcome of rehospitalization, emergency department visits, or death. This shows a high discrepany in findings and is probably due to the medical attributes of the study sample: other studies usually target high-risk patients whereas in DeBusk et al intervention was for a sample of HMO enrollees out of 5 medical centres in the northern California region, representative of 18 Kaiser Permanente facilities. The authors concluded that a specialized program for patients with mild HF may increase health costs without improving clinical outcomes, such as rehospitalization.

A summary based on a Cochrane review of 14 STS {16} was performed. All-cause hospitalization data were available for 11 STS studies, and STS was effective in reducing the rate of all-cause hospitalization in patients (p<0,02). STS was effective in reducing the hospitalizations for CHF (p<0,0001).One of six studies was significant in terms of reduction in length of stay. Nine studies presented costs for STS – the costs varied accordingly to intensity and technologies used. Studies that reported cost reduction in the cost of care per admission or overall costs reduction due to reduction in hospitalization reported cost savings ranging between 35% and 86%.

Klersy et al {1430} performed a meta analysis of 21 RCTs (5.715 patients). RM was associated with a significantly lower number of hospitalizations for HF, while LOS was not different. Direct costs for hospitalization for HF were approximated by DRG tariffs in Europe and North America. The difference in costs between RM and UC ranged from EUR 300 and EUR 1.000, favouring RM. Estimated costs ranged from 991,2 EUR to 3.207,96 EUR annually, depending on DRG reimbursement rate (from EUR 2.360 to EUR 7.638). A simulation study was performed on a hypothetical cohort of 100 patients observed for 1 year. The initial hypothesis assumed that all patients were treated to UC; then the proportion of these patients followed using an RM management strategy was progressively increased up to 50%. Due to a difference in incidence rate (assumed 0,42 for UC and 0,29 for RM) the adoption of RM strategy entailed a progressive and linear increase in payer/system costs saved. The study concludes that management of HF patients by RM is costs saving driven by a reduction in the number of HF hospitalization. No other costs (like costs of RM etc) or effects (less travel etc) are taken in to account. Authors conclude that the economic data collected in RCTs are scanty.

Analysis of total healthcare utilization as well as CHF-related healthcare utilization {2} showed that in no area, including drug use, office visits, emergency department visits, procedures, or hospitalizations, was there a decrease with the intervention. Moreover, total healthcare cost was not statistically different between groups. Total costs per patient were computed by summing the estimated costs for the 5 resource categories by the three 6-month study periods, and then summing these 3 totals.The costs of administering the intervention were not included. Total costs for 6 months amounted to $3.001,26 for control group and $3.277,05 for intervention group. The difference was not statistically significant. Galbreath et al {2} strongly suggests that widespread application of DM programs for all patients with CHF may have limited impact on healthcare costs. The lack of costs difference persisted even in patients with higher NYHA class, suggesting that the improved survival in sicker patients was not associated with cost savings.

DM will be most useful when there is a marked disparity between the best case management for a disease and the management that is being applied. Earlier DM trials were performed when use of guideline-based therapy was less prevalent that it is currently. In this study {2} 77% of patients were on an ACE inhibitor at the time of enrollment. The startling reductions in hospitalizations and healthcare costs reported by some earlier investigators may not be achievable at present. DM will be most useful when targeted to patients with worse functional class.

One of the latest study by Grustam et al {310} included 15 studies that described a telephone case management. More than 70 % of the studies did not take into account expenses in one of the following categories; healthcare sector, other sectors, patient/family expenses or productivity losses. As a positive example, Wennberg et al (25) estimated the total costs and not just the marginal cost: they included salaries and benefits, training expenses, amortized capital expenditures, data and coaching operations, fulfillment and overhead. None of the studies analyzed a shift of costs from specialist physicians to HF nurse to GP, for instance. In 80% of the studies the source and methods of the evaluations were not clear. Authors mostly focused on direct costs while omitting indirect and intangible costs. Principally, the costs were missing across majority of the studies and those of the intervention overheads, training of personnel, and patient related costs.  

Critical
Partially
Rupel V et al. Result Card ORG8 In: Rupel V et al. Organisational aspects In: Jefferson T, Cerbo M, Vicari N [eds.]. Structured telephone support (STS) for adult patients with chronic heart failure [Core HTA], Agenas - Agenzia nazionale per i servizi sanitari regionali ; 2015. [cited 16 June 2021]. Available from: http://corehta.info/ViewCover.aspx?id=305

References