An evaluation of telehealth expansion in U.S. nursing homes

Abstract

Objective

This research brief contains results from a national survey about telehealth use reported in a random sample of U.S. nursing homes.

Methods and Materials

The sample includes nursing homes (N = 664) that completed surveys about information technology maturity, including telehealth use, beginning January 1, 2019, and ending August 4, 2020. A pre/post design was employed to examine differences in nursing home telehealth use for nursing homes completing surveys prior to and after telehealth expansion, on March 6, 2020. We calculated a cumulative telehealth score using survey data from 6 questions about extent of nursing home telehealth use (score range 0-42). We calculated proportions of nursing homes using telehealth and used logistic regression to look for differences in nursing homes based on organizational characteristics and odds ratios.

Results

Significant relationships were found between nursing home characteristics and telehealth use, and specifically, larger metropolitan homes reported greater telehealth use. Ownership had little effect on telehealth use. Nursing homes postexpansion used telehealth applications for resident evaluation 11.24 times more (P < .01) than did nursing homes pre-expansion.

Discussion

Administrators completing our survey reported a wide range of telehealth use, including approximately 16% having no telehealth use and 5% having the maximum amount of telehealth use. Mean telehealth use scores reported by the majority of these nursing homes is on the lower end of the range.

Conclusions

One solution for the current pandemic is to encourage the proliferation of telehealth with continued relaxed regulations, which can reduce isolation and preserve limited resources (eg, personal protective equipment) while maintaining proper distancing parameters.

INTRODUCTION

The unprecedented coronavirus disease 2019 (COVID-19) global virus pandemic has left a wake of uncertainty for nursing home providers (ie, doctors, nurse practitioners, clinical psychologists, and licensed clinical social workers), residents, and families of residents who are isolated to mitigate exposure. This isolation was necessary to slow the spread of the virus and protect some of the most vulnerable populations. In response to the crisis, Medicare rules and regulations have been relaxed to broaden access to telehealth services in nursing home settings so that Medicare beneficiaries receive the care they need virtually without having to risk travel to a healthcare facility.¹ We hypothesize that relaxation of federal regulations with telehealth expansion will lead to greater nursing home telehealth adoption nationally, a setting that accounts for over 10 000 (27%) of COVID-19 deaths in the United States.²

In response to telehealth expansion, the Centers for Medicare and Medicaid Services published a general provider telehealth toolkit to assist healthcare facilities plan for implementation of these types of services including Medicare telehealth visits, virtual check-in, and e-visits (electronic visits).¹ As of April 20, 2020, every U.S. state has amended existing laws or issued new declarations to expand the use of telehealth or mHealth (mobile health) during the COVID-19 pandemic.³ These amendments to existing laws provide opportunities for the use of telehealth, that is, using technology to assist with procedures, such as medical screenings, consultations, and second opinions, and to prevent contact with contagious people during the outbreak, a method thought to protect patients and healthcare providers.⁴

The Centers for Medicare and Medicaid Services toolkit specifically defines telehealth visits as using telecommunication to conduct visits between a provider and resident. Some examples include a virtual check-in, a short 5- to 10-minute visit or remote evaluation with a provider via a telephone or other telecommunication device (ie, Skype, Facetime, or Zoom) to determine what services are required for the resident and e-visits, and any communication occurring between a provider and resident using an online portal. In this study, we conceptualize these tools as forms of telehealth used by nursing homes to protect people living and working in these facilities. To remove potential delays in implementation, the Health and Human Services Office of Civil Rights also exercised enforcement discretion and waived Health Insurance Portability and Accountability Act violations for providers using telehealth tools in the care of these vulnerable populations, which could also result in greater levels of telehealth expansion.¹

The purpose of this article is to examine the use of telehealth services being reported in a random sample of U.S. nursing homes. Telehealth is a valuable resource identified in the nursing home literature, as it has been associated with reducing polypharmacy, exploring treatable causes of weight gain, and reducing frailty.⁵ Telehealth applications can now be viewed in a broader sense to help reduce or eliminate infections caused by known contagious organisms spread through various means of contact. Hence, demonstrating greater significance of this research to understand pre- and postexpansion rates of telehealth usage following the public health emergency declaration in nursing homes nationally. Furthermore, not all nursing homes are alike. Accounting for organizational characteristics that are used frequently in nursing home research involving technology⁶ will help to describe potential disparities in telehealth implementation. Our research questions are the following:

1. Did telehealth expansion, beginning March 6, 2020, increase telehealth uptake in U.S. nursing homes?

2. What are nursing homes using telehealth for?

3. Are there differences in nursing home telehealth uptake pre– and post–telehealth expansion based on facility characteristics (bed size [<60, 60-120, >120], location [metropolitan, micropolitan, small town, rural], and ownership [for profit, not for profit])?

MATERIALS AND METHODS

Sample

Facilities were randomly selected from Nursing Home Compare⁷ after removing facilities from Guam, Puerto Rico, and Virgin Islands. Facilities that were designated as special focus facilities and that identified as having poor quality outcomes or safety issues requiring focused interventions to fix problems were also removed. Included in this analysis are survey results from a random sample of nursing homes that completed surveys beginning January 1, 2019, and ending August 4, 2020, which provides an estimate of pre– and post–telehealth expansion telehealth use, based on the March 6, 2020, expansion date.

Survey items

A national survey began January 1, 2019, in U.S. nursing homes to examine trends in information technology (IT) maturity over 3 years. As part of this evaluation, researchers are using a nursing home survey that measures the extent of telehealth use in resident care and clinical support domains (eg, laboratory, radiology, pharmacy).⁸ A detailed description of the survey has been previously published.⁹ The survey has been tested and determined to have good reliability and validity measures.^10,11

To answer our research questions, we focused on 6 telehealth questions in the survey (see Table 1). These 6 questions measure extent of use of telehealth applications, including those used to facilitate medical screenings, conduct follow-up visits and consultations, and perform medication management activities virtually, thus reducing unnecessary exposure to threatening environmental agents, in resident care and clinical support areas (eg, laboratory, radiology, pharmacy). In this context, we consider all these applications as telehealth activities.

Table 1.

Telehealth survey questions with proportions of telehealth use (N = 664)

Survey question	Yes	No
1. Telehealth for evaluation of residents and pretransfer arrangements	250 (37.65)	414 (62.35)
2. Telehealth for transmission of diagnostic images and/or consultations and second opinions	257 (38.70)	407 (61.30)
3. Electronic reporting of laboratory test results to nursing home	442 (66.57)	222 (33.43)
4. Electronic transmission and reception of laboratory results for interpretation (eg, pathology)	393 (59.19)	271 (40.81)
5. Telehealth for results capturing and interpretation by radiologists	184 (27.71)	480 (72.29)
6. Remote order entry for medications from locations outside of the nursing home (eg, MD access from home, office or clinic)	355 (53.46)	309 (46.54)

Questions scored on an 8-point scale ranging from 0 (Not Available) to 7 (Extensively Used).⁹

Participants were asked to rate these survey items according to their telehealth extent of use using an 8-point scale ranging from 0 (not available) to 7 (extensively used). We calculated a cumulative telehealth score using data from questions 1-6 for each home, with a minimum score of 0 and maximum of 42. We created binary outcome measures for these variables by assigning “0” if the respondent indicated 0 and “1” if the respondent indicated between 1 and 7. Subsequently, we calculated proportions of telehealth use in nursing homes completing the survey.

Analysis

We used descriptive statistics to compare our study sample with the national sample relative to ownership, bed size, and location. These specific organizational variables were selected because they were available in a public dataset that all nursing homes receiving Medicare are required to report.¹² In addition, these variables were selected because they were found to have significance in previous analyses conducted to evaluate relationships between IT maturity and quality measures, also found in Nursing Home Compare.^10,13 Next, we incorporated poststratification weighting procedures to reweight the nursing homes to national proportions regarding these variables. Using poststratified weights, the team looked at differences in total telehealth use scores among each of the 6 survey questions contributing to the total telehealth use score.

Next, we used logistic regression for survey data to examine the relationship between each type of telehealth use, based on the 6 survey questions, and nursing home characteristics including location (rural, small town, micropolitan, metropolitan), bed size (<60, 60-120, >120), and type of ownership (for profit, not for profit). Location was determined using rural-urban commuting area codes established from census data population statistics (rural = <2500, small town = 2500-9999, micropolitan = 10 000-49 999, metropolitan = >50 000).¹⁴ Finally, with poststratified data, we assessed relationships between telehealth use in nursing homes completing surveys prior to and after telehealth expansion on March 6, 2020, by incorporating an additional variable into the logistic regression models to calculate odds ratios (OR) and 95% confidence intervals while adjusting for nursing home characteristics.

RESULTS

The nursing home sample (n = 664) was reflective of the population (N = 13 958) according to location but not according to bed size or ownership. This sample had a greater proportion of smaller (<60 beds) and medium-sized (60-120 beds) nursing homes but had fewer large nursing homes (>120 beds) compared with the national sample. The sample also had a larger proportion of nonprofit facilities.

Nursing home administrators in our study reported a full range of telehealth use scores ranging from 0 to 42. Among this sample (n = 664), 105 (16%) nursing homes had a telehealth use score of 0 and 32 (5%) nursing homes had the maximum telehealth use score (42). Partial telehealth implementations were reported in 527 (79%) nursing homes, the majority being on the lower end of the distribution. See Figure 1 for distribution of telehealth use scores.

Figure 1.

Distribution of total telehealth use score (N = 664).

Table 2 shows mean telehealth use scores according to nursing home characteristics of bed size, ownership, and location. In this sample, mean telehealth use scores increased as bed size increased from small (<60) to large (>120). Mean telehealth use scores were lower in nursing homes in rural locations compared with in nursing homes located where populations sizes were much larger. Although our sample included many more for-profit nursing homes, mean telehealth use scores and confidence intervals were not much different based on ownership status. The mean telehealth use score reported by facility administrators during the pre-expansion period were lower than during the post-expansion period, indicating greater telehealth adoption overall during the postexpansion period when telehealth regulations were relaxed.

Table 2.

Table 2. Telehealth use score by nursing home characteristics and survey period

Nursing home characteristics	Total facilities	Telehealth use score
Bed size
<60	139	11.45 (9.51-13.40)
60-120	388	15.70 (14.47-16.94)
>120	137	16.39 (14.57-18.22)
Ownership
For profit	480	15.27 (14.14-16.39)
Not for profit	184	14.63 (13.05-16.21)
Location (population)
Metro (50 000)	428	15.82 (14.66-16.98)
Micro (10 000-49 999)	99	15.23 (12.70-17.77)
Small town (2500-9999)	82	13.54 (11.06-16.02)
Rural (<2500)	55	10.82 (8.64-13.01)
Survey period
Pre–telehealth expansiona	491	13.51 (12.49-14.52)
Post–telehealth expansion	173	19.08 (17.03-21.12)

Values are mean (95% confidence interval).

^aSurvey period prior to March 6, 2020.

Table 3 illustrates the relationships between telehealth use score, organizational characteristics, and survey period, including change over time between pre–telehealth expansion (prior to March 6, 2020) and post–telehealth expansion. Using poststratification to correct for sample differences, there were significant associations found among the dependent variables and organizational characteristics in our logistic regression analysis (Table 3). For instance, medium-sized (60-120 beds) and larger nursing homes (>120 beds) were 1.85 (P = .01) and 2.89 (P < .01) times more likely to have electronic reporting of lab results to the nursing home compared with smaller facilities (<60 beds), respectively. In addition, nursing homes located in micropolitan and metropolitan areas, with larger populations, were 2.30 (P = .04) and 3.44 (P < .01) times more likely to have electronic reporting of laboratory results compared with nursing homes in rural locations, respectively. Similar results were found for electronic transmission and reception of laboratory results with larger (>120 beds) and medium-sized (60-120 beds) nursing homes having 2.10 (P < .01) and 1.83 (P < .01) times the odds of having this telehealth capability compared with small homes (<60 beds), respectively. Also, metropolitan nursing homes were 3.19 times more likely to have transmission and reception of laboratory results capability compared with rural nursing homes (P < .01). Telehealth used for results capturing and interpretation by radiologists were more likely to be used in nursing homes in larger metropolitan areas (OR, 2.86; P < .01) and small towns (OR, 2.95; P = .03) compared with rural nursing homes. Finally, the odds of a nursing home having remote order entry capabilities for medications was greater in medium-sized (OR; 2.16; P < .01) and larger (OR, 2.02; P = .01) facilities.

Table 3.

Relationships of telehealth use score, organizational characteristics (bed size, location, ownership), and survey period

Dependent variable	Independent variable	Parameter estimate	Odds ratio (95% CI)	P value
Telehealth for evaluation of residents and pretransfer arrangements	Bed size
	<60	Ref	Ref	Ref
	60-120	−0.02	0.98 (0.60-1.62)	.95
	>120	−0.51	0.60 (0.31-1.17)	.14
	Location
	Metro	0.12	1.13 (0.50-2.58)	.77
	Micro	0.52	1.69 (0.65-4.40)	.29
	Small town	0.12	1.13 (0.42-3.01)	.81
	Rural	Ref	Ref	Ref
	Ownership
	For profit	0.35	1.42 (0.90-2.25)	.13
	Nonprofit	Ref	Ref	Ref
	Survey period
	Pre–telehealth expansion	Ref	Ref	Ref
	Post–telehealth expansion	2.42	11.24 (7.21-17.53)	<.01
Telehealth for transmission of diagnostic images and/or consultations and second opinions	Bed size
	<60	Ref	Ref	Ref
	60-120	0.18	1.20 (0.75-1.92)	.44
	>120	−0.15	0.86 (0.48-1.55)	.61
	Location
	Metro	−0.30	0.43 (0.36-1.53)	.42
	Micro	0.07	1.07 (0.45-2.58)	.87
	Small town	−0.44	0.64 (0.28-1.50)	.31
	Rural	Ref	Ref	Ref
	Ownership
	For profit	0.23	1.25 (0.82-1.92)	.30
	Nonprofit	Ref	Ref	Ref
	Survey period
	Pre–telehealth expansion	Ref	Ref	Ref
	Post–telehealth expansion	1.46	4.30 (2.85-6.51)	<.01
Electronic reporting of laboratory test results to nursing home	Bed size
	<60	Ref	Ref	Ref
	60-120	0.62	1.85 (1.15-2.97)	.01
	>120	1.06	2.89 (1.55-5.38)	<.01
	Location
	Metro	1.24	3.44 (1.71-6.91)	<.01
	Micro	0.83	2.30 (1.04-5.11)	.04
	Small town	0.57	1.76 (0.77-4.00)	.18
	Rural	Ref	Ref	Ref
	Ownership
	For profit	−0.27	0.77 (0.48-1.21)	.25
	Nonprofit	Ref	Ref	Ref
	Survey period
	Pre–telehealth expansion	Ref	Ref	Ref
	Post–telehealth expansion	0.17	1.19 (0.77, 1.83)	.44
Electronic transmission and reception of laboratory results for interpretation (eg, pathology)	Bed size
	<60	Ref	Ref	Ref
	60-120	0.61	1.83 (1.21-2.78)	<.01
	>120	0.74	2.10 (1.26-3.52)	<.01
	Location
	Metro	1.16	3.19 (1.72-5.92)	<.01
	Micro	0.52	1.69 (0.84-3.40)	.14
	Small town	0.33	1.39 (0.67-2.86)	.38
	Rural	Ref	Ref	Ref
	Ownership
	For profit	−0.38	0.68 (0.47-0.99)	.05
	Nonprofit	Ref	Ref	Ref
	Survey period
	Pre–telehealth expansion	Ref	Ref	Ref
	Post–telehealth expansion	0.23	1.26 (0.87, 1.82)	.22
Telehealth for results capturing and interpretation by radiologists	Bed size
	<60	Ref	Ref	Ref
	60-120	0.22	1.25 (0.75-2.07)	.40
	>120	0.19	1.21 (0.65-2.25)	.56
	Location
	Metro	1.05	2.86 (1.25-6.50)	.01
	Micro	0.89	2.45 (0.93-6.40)	.07
	Small town	1.08	2.95 (1.14-7.66)	.03
	Rural	Ref	Ref	Ref
	Ownership
	For profit	0.11	1.12 (0.73-1.72)	.60
	Nonprofit	Ref	Ref	Ref
	Survey period
	Pre–telehealth expansion	Ref	Ref	Ref
	Post–telehealth expansion	1.06	2.89 (1.89-4.42)	<.01
Remote order entry for medications from locations outside of nursing home (eg, MD access from home, office, or clinic)	Bed size
	<60	Ref	Ref	Ref
	60-120	0.77	2.16 (1.36-3.42)	<.01
	>120	0.70	2.02 (1.15-3.54)	.01
	Location
	Metro	0.45	1.56 (0.82-2.98)	.17
	Micro	0.21	1.23 (0.58-2.62)	.59
	Small town	0.01	1.01 (0.45-2.24)	.98
	Rural	Ref	Ref	Ref
	Ownership
	For profit	−0.05	0.95 (0.63-1.43)	.81
	Nonprofit	Ref	Ref	Ref
	Survey period
	Pre–telehealth expansion	Ref	Ref	Ref
	Post–telehealth expansion	0.34	1.41 (0.95-2.10)	.09

CI, confidence interval; Ref, Reference Variable.

Adjusting for bed size, location, and ownership, Table 3 illustrates the comparisons of telehealth use pre– and post–telehealth expansion based on surveys completed before and after March 6, 2020. The regression table shows statistically significant differences in mean telehealth use scores among 3 areas, including telehealth used for evaluation of residents and pretransfer arrangements, transmission of diagnostic images or consultations and second opinions, and results capturing and interpretation by radiologists. In particular, nursing homes in the postexpansion period were 11.24 times more likely (P < .01) to use telehealth for resident evaluation and pretransfer arrangements compared with facilities in the pre-expansion period. Nursing homes in the postexpansion period had 4.30 times the odds of using telehealth for consultations and second opinions vs facilities in pre-expansion period. Finally, in this sample, radiologists were 2.89 times more likely to use telehealth for results capturing and interpretation after the telehealth expansion (P < .01). Nursing home administrators did not report significant changes in other telehealth opportunities after expansion occurred on March 6, 2020.

DISCUSSION

Administrators completing our survey reported a wide range of telehealth use including approximately 16% having no telehealth use, 5% having the maximum amount of telehealth use, and 79% reporting partial telehealth implementations. Our finding that facilities have partial technology implementations especially in clinical support areas, such as laboratory systems and pharmacy systems, are consistent with other research in this area.¹⁵ Findings from this study suggest that there are gaps and opportunities in the use of telehealth, such as opportunities for building greater interoperability among telehealth systems supporting providers decision-making abilities and enhanced provider order entry, which could improve timeliness and safety of care delivery in nursing homes.

The overall mean telehealth use scores reported by the majority of these nursing homes is on the lower end of the range, which indicates that there is much room for improvement. These findings are supported because adoption of newer forms of technology have struggled to achieve a maximum adoption level. Some reasons nursing home administrators struggle include a need for systematic implementation processes and evidenced based protocols, lack of technology support and infrastructure, low levels of interoperability among disparate systems, and poor investments in staff training.¹⁶ However, there are bright spots, with several homes, who likely have strong leadership advocating for technology, having adopted high levels of telehealth use.¹⁷ The relaxation of current regulations and the formation of telehealth toolkits due to the COVID 19 pandemic maybe useful, but it remains to be seen if loosening regulations can help nursing home administrators overcome some of the monumental struggles that they have experienced trying to keep pace with other health sectors (eg, acute care) who have traditionally been provided greater financial resources for technology implementation.¹⁸

This analysis illustrates telehealth use has some significant relationships with nursing home location and size but less significant relationships with type of ownership. Telehealth that is associated with the exchange of laboratory results, results capturing and reporting by radiologists, and remote order entry by pharmacists for medications were significant in our study. In prior studies, research has shown that smaller and more rural nursing homes are often found to have greater disparities in technology use compared with facilities in larger, urban areas.¹⁹ One concern about these findings is whether access to telehealth in larger, urbanized nursing homes will lead to greater disparities among vulnerable populations in rural locations. This is especially important during pandemic times, when isolation precautions and social distancing are required to protect vulnerable residents and staff who work in these facilities. Only time and ongoing research on this topic will tell.

LIMITATIONS

Recruitment for this study was grouped according to state. Because we had only begun recruiting nursing homes in some states for our national survey, some states had fewer strata represented, especially for larger, nonprofit nursing homes. As we complete our national assessment, scheduled to be completed in September 2, 2020, we anticipate that our randomization process will correct for this lack of representation.

CONCLUSION

Telehealth technology is thought to be a critical access point to health care for vulnerable populations, chronically ill nursing home residents, and people living in rural settings.²⁰ The current situation for most nursing home residents, staff, and administrators worldwide, as a result of the COVID-19 pandemic, includes greater isolation and separation to prevent spread of the virus. Without a vaccine and electronic connections to the outside world, nursing home residents could spend weeks, months, or even years in a facility without visitors other than regular staff. This could have a profound effect on resident outcomes including depression rates, mobility, etc. One solution is to encourage the proliferation of telehealth with continued relaxed regulations that can reduce isolation and preserve limited resources (eg, personal protective equipment) while maintaining proper distancing parameters and allowing for timely care delivery and social connectedness everywhere.

FUNDING

This project was supported by grant number R01HS022497 from the Agency for Healthcare Research and Quality. The content is solely the responsibility of the authors and does not necessarily represent the official views of the Agency for Healthcare Research and Quality.

AUTHOR CONTRIBUTIONS

GLA, CBD, and KRP all provided substantial contributions to the over design, acquisition, analysis, and interpretation of the data for this manuscript. GLA, CBD, and KRP assisted in original drafts and revisions of the manuscript. All authors gave final approval of the last version submitted for publication. All authors agree to be accountable for all aspects of the work, including but not limited to accuracy and integrity of statements in this publication.

CONFLICT OF INTEREST STATEMENT

GLA is owner and cofounder of TechNHOlytics LLC. GLA is also a member of the Agency for Healthcare Research and Quality National Advisory Committee.