An estimated 5 million Americans have congestive heart failure (CHF) and one in five over the age of 40 will develop CHF. There are numerous examples of CHF patients living beyond the years normally expected for people with the disease, usually attributed to taking an active role in disease management. A relatively new alternative for CHF outpatient care is telemedicine and e-health. We investigated the effects of a 6-week in-home telemedicine education and monitoring program for those with systolic dysfunction on the utilization of health care resources. We also measured the effects of the unit 4.5 months after its removal (a total of 6 months post introduction of the unit into the home). Concurrently, we assessed participants' perceptions of the value of having a telemedicine unit. Participants in the telemedicine group reported weighing more times a week with less variability than did the control group. Telemedicine led to a reduction in physician and emergency department visits and those in the experimental group reported the unit facilitating self-care, though this was not significantly different from the control group (possibly due to small sample size). These findings suggest a possibility for improvement in control of CHF when telemedicine is implemented. Our review of the literature also supports the role of telemedicine in facilitating home health care and self-management for CHF patients. There are many challenges still to be addressed before this potential can be reached and further research is needed to identify opportunities in telemedicine.
The American Heart Association reports that
one in five Americans over the age of 40 will develop congestive heart failure
(CHF) [
CHF is a chronic condition
where appropriate disease management is critical [
Experts agree that
outpatient care is important in achieving the best possible outcomes for
patients with CHF, viable forms are currently being investigated and debated
[
Further, the potential for
telemedicine and e-health is particularly promising for those living in rural
areas where health care access is diminished [
This study investigated the efficacy of a 6-week in-home telemedicine monitoring program, as well as measured whether the program would retain effectiveness during a 6-month period by comparing health care utilization scores. The utilization of health care was determined by the number of times participants contacted and visited physicians, emergency department visits, and hospitalizations. The patients were also assessed based on the number of times they weighed themselves per week, and NYHA scores pre and post experimental period. In addition, participants' perceptions of the value of having a telemedicine unit were also obtained.
Patients
who met the criteria (refer to Table
Selection criteria.
Documented CHF due to systolic dysfunction | CHF was caused by diastolic dysfunction |
Is on renal dialysis | |
CHF with a New York heart classification (NYHC) of II–IV | Has dementia or another uncontrolled psychiatric disorder that can interfere with his/her ability to participate |
Documented EF (ejection fraction) <40% by echo, nuclear medicine, or CCL within 6 months of enrollment | Anticipated survival from a non-CHF cause is less than 6 months |
Can read and speak English | Participated in another heart failure research protocol within the previous 6 months |
Has an active phone line in his/her home | Currently receives home health nursing services |
Has had a heart transplant | |
Is able to give his/her own informed consent | If pregnant or of child-bearing age and is trying to become pregnant |
Lives at home; is 18 years or older | Is blind, unable to use his/her upper extremities, or has any physical condition that may inhibit him/her from viewing and/or using a computer screen |
Participants
in the telemedicine group were instructed to complete a monitoring session each
morning for the 6-week period. A daily
monitoring session included weighing on
a scale and answering a health questionnaire. If there was a weight gain or
loss of
A chart review was conducted and a phone interview was completed by the research team 6 months from initial start date. Information was obtained regarding the number of times the patient contacted the physician's office, was seen by the physician, was seen in the emergency department, and/or admitted to the hospital. Further information was obtained regarding the number of times a patient weighed themselves a week.
The telemedicine unit was ViTel Net's DataGate system, which included a notebook PC, a touch-screen monitor, and a scale. The notebook PC dimensions were approximately 9.3 inches by 6.9 inches by 2 inches. The monitor dimensions were approximately 12 inches by 9 inches by 3 inches. The A&D LifeSource MD Digital Scale UC-321PL scale was used to measure the weight in 1/10 pound increments, up to a maximum of 450 pounds. The weight value was sent via the RS232 port.
The PC unit connected to a standard electrical wall outlet. The system was very user friendly, including a 12 inch color touch screen with simple point-and-click buttons for presenting educational material, audio prompts, and custom workflow sequences (ViTel Net Description). Once the telemedicine unit was installed in the participant's home, they received the telemedicine participant information packet, which contained instructions for use and care of the equipment, a troubleshooting guide, and research contact information (names and phone numbers) for the research nurse, principal investigator, and project manager. The participant received instruction on the equipment and then was asked to demonstrate appropriate use of the equipment to complete the training session.
Of the 23 participants, 13
people received the telemedicine program (age 50–81, mean 66.08; 8 male, 5
female; 69% married; 100% Caucasian) and 10 people (age 55–90, mean 71.00; 8
male, 2 female; 80% married; 100% Caucasian) were in the control group. As was our intention, the two groups were similar
in age, sex, marital status, years of education, and NYHA classification scores (refer to Table
Participant information (means).
Telemedicine | Control | |
---|---|---|
Age | ||
Years of education | ||
Martial status | ||
Single | 2 | 0 |
Married | 9 | 8 |
Divorced | 1 | 0 |
Widowed | 1 | 2 |
New York heart classification, start of study | ||
New York heart classification, end of study |
We compared telemedicine and control group health
care utilization scores using t-tests and chi-square
according to variable type and
found no statistical significance.
Descriptive statistic analysis revealed a pattern of data that suggests
the possibility of reaching statistical significance if the sample size were
greater. For example, examination of
percentages of contacts to physician, physician visits, and emergency
department visits conveyed a pattern of higher scores (though not statistically
reliable) for the control group (refer to Figure
Control and telemedicine patient utilization of healthcare services.
Contact physician | Number of visits | Percentage | Number of visits | Percentage |
0 | 37.5 Percentage | 0 | 38.5 | |
1 | 25.0 Percentage | 1 | 38.5 | |
2 | 12.5 Percentage | 4 | 7.7 | |
3 | 12.5 Percentage | 5 | 7.7 | |
13 | 12.5 Percentage | 9 | 7.7 | |
Visit physician | Number of visits | Percentage | Number of visits | Percentage |
1 | 12.5 Percentage | 0 | 16.7 Percentage | |
2 | 37.5 Percentage | 1 | 25.0 Percentage | |
3 | 12.5 Percentage | 2 | 33.3 Percentage | |
4 | 12.5 Percentage | 3 | 8.3 Percentage | |
6 | 12.5 Percentage | 5 | 8.3 Percentage | |
9 | 12.5 Percentage | 7 | 8.3 Percentage | |
Emergency department | Number of visits | Percentage | Number of visits | Percentage |
1 | 28.6 Percentage | 1 | 15.4 Percentage | |
Hospitalized | Number of times | Percentage | Number of times | Percentage |
2 | 14.3 Percentage | 1 | 23.1 Percentage | |
— | — | 2 | 23.1 Percentage | |
Weigh per week | Mean = 3.81 | Mean = 4.94 | ||
Standard deviation = 3.69 | Standard deviation = 2.31 |
Percent of patients who contacted their physician (CP), visited their physician (PV), utilized the emergency department (ED), and/or were hospitalized (Hos).
We analyzed information
regarding the experience of being in the telemedicine group and found that
despite nearly half (46%) of the participants reporting no previous experience
with computing systems, all but one of the participant reports were positive. The person who did not have a positive
experience indicated that “the telemedicine unit made her worry too much about
her condition so it stressed her heart.”
When asked the open-ended question of what participants liked best about
having the telemedicine unit in their home, the majority indicated that the
unit helped them be aware of factors important for managing their disease and
helped them to control and be more aware of their weight. Seventy-five percent of participants
indicated they would like to continue using the unit. Though we are unsure why
25% did not wish to continue, some reported complaints were
“repetitive lessons,” “redundant questions,” and
“would be more effective if interaction was involved.” The data analysis revealed numerous ways by which the
telemedicine unit facilitated self-care. Spearman's rho correlation revealed
significant correlates between improved self-care and usage of the telemedicine
unit. For example, those who believed that having the telemedicine unit in their home will help them take better care of themselves in the future were highly correlated with better understanding their
conditions,
Facilitation of self-care correlates.
Helped take better care of self | Liked unit in home | 0.752 |
Will help take care of self in the future | 0.887 | |
Better understand condition | 0.665 | |
Recommend telemedicine unit | 0.795 | |
Taking medications daily | 0.827 | |
Important symptoms | 0.760 | |
Use unit again | 0.751 | |
Will help take care of self in the future | Liked unit in home | 0.911 |
Better understand condition | 0.854 | |
Recommend telemedicine unit | 0.924 | |
Limit salt | 0.712 | |
Taking medications daily | 0.934 | |
Important symptoms | 0.753 | |
Call to seek help | 0.739 | |
Stay healthier | 0.753 | |
Use unit again | 0.777 | |
Better understand condition | Liked unit in home | 0.722 |
Recommend telemedicine unit | 0.928 | |
Limit salt | 0.762 | |
Taking medications daily | 0.925 | |
Important symptoms | 0.622 | |
Call to seek help | 0.868 | |
Stay healthier | 0.870 | |
Use unit again | 0.707 |
Correlates: perspectives regarding the telemedicine unit.
Liked unit in home | Helped take better care of self | 0.752 |
Satisfied with training | 0.738 | |
Will help take care of self in the future | 0.911 | |
Better understand condition | 0.722 | |
Recommend unit | 0.890 | |
Limit salt | 0.648 | |
Taking medications daily | 0.891 | |
Important symptoms | 0.775 | |
Call/seek help | 0.771 | |
Stay healthier | 0.724 | |
Use unit again | 0.818 | |
Recommend unit | Helped take better care of self | 0.795 |
Will help take care of self in the future | 0.924 | |
Better understand condition | 0.928 | |
Limit salt | 0.741 | |
Taking medications daily | 0.999 | |
Important symptoms | 0.714 | |
Call/seek help | 0.852 | |
Stay healthier | 0.855 | |
Use unit again | 0.811 |
Despite finding no statistically significant results when comparing the experimental and control
groups for health care utilization rates, we are encouraged by the reports from
those in the telemedicine group indicating the unit facilitated self-care
(refer to Figure
Percent of patients' overall satisfaction with the telemedicine unit and satisfaction in the unit's assistance in managing CHF.
The inclusion/exclusion criteria used in this study resulted in a sample that was probably too small to glean statistically reliable results. For example, we had 105 referrals for participation. Of those, 23 enrolled, 61 were excluded because of not meeting inclusion criteria, and 21 were deceased either before enrollment or before completing the study. Numerous others who would have met the inclusion/exclusion criteria were too ill to participate. Of the 23 enrolled in the study, two were deceased before the study period ended and two more died within a month of ending the study. Of the 61 who did not meet inclusion criteria, most were due to an EF of greater than 40%, CHF was not due to systolic dysfunction, or did not have an echo within six months.
This study was conducted in Idaho, USA, which is considered a rural state and where we hope to better promote the use of telemedicine in rural communities. Though telemedicine seeks to benefit this demographic, distance, and physical disabilities complicate travel, and to hinder research efforts. Due to software limitations, only English speaking patients were included. Additionally, our study consisted of only white participants. Although Idaho is predominately English speaking and white, an accurate model would include other ethnicities.
The largest challenge in many telemedicine studies is obtaining study participants. The difficulties that are achieving adequate enrollment for statistically reliable comparisons indicate that future research designs may benefit from less stringent inclusion and exclusion criteria. Obtaining more patients in less advanced stages of a disease may be one way to improve enrollment. People in all stages of a chronic condition may benefit from home monitoring and increased education. Comparisons of telemedicine's effects at various stages in the disease process may improve data on the efficacy of telemedicine. For example, people in the early stages of a disease may particularly benefit from an early telemedicine intervention, as proper self-care may slow the disease process.
A central goal of telemedicine is to improve delivery of health services to underserved areas. However, enrollment difficulties are exacerbated in studies focusing on rural residents. Chronic conditions inhibit travel and mobility for all patients, regardless of residency location. Therefore, expansion of study demographics to include both urban and rural participants may be another way to facilitate improved enrollment. Future research can also be strengthened with the addition of a multilingual version of the software. This will prevent any biases from restricting the study population. In a future home monitoring project, regarding diabetes management, we will use upgraded equipment from the current study and plan to work with clinics in both rural and urban areas to obtain a larger sample size. This study will provide an opportunity to examine differences and similarities between rural and urban populations in the health benefits achieved through the use of telemedicine.
Having decreased hospitalization as a primary metric and goal may not accurately convey efficacy of telemedicine. It is possible that greater awareness prevented delays in necessary hospital admission; thus, better responsiveness occurred. If so, this would be a positive factor rather than an indication that the telemedicine program was unsuccessful. We recommend this issue to be considered in future research designs.
While home monitoring and education for chronic conditions facilitates improved self-care, people still need increased access to specialty services. In many cases, people in rural areas or people with decreased mobility may not receive care that could improve health. Telemedicine applications that allow specialists to view and communicate with patients remotely may help improve outcomes. Hybrid models of health care delivery, including home monitoring and education, telemedicine access to specialists, and in-person visits may improve patient care and outcomes and help alleviate the problems associated with physician shortages.
The authors would like to thank and recognize the Telemedicine and Advanced Technology Research Center, Department of Defense for a grant supporting this research through Award no. WB1XWH-04-1-0604. The U.S. Army Medical Research Acquisition Activity, Fort Detrick MD 21702-5014, was the awarding and administering acquisition office. The content and information included does not necessarily reflect the position or the policy of the Government, and no official endorsement should be inferred.