Efficacy of Telehealth in Preoperative Pediatric Surgery Consultations

Abstract

Introduction

The Coronavirus disease 2019 pandemic created a sudden need to transition outpatient pediatric surgical care to a telehealth platform, allotting little time to study the efficacy of these changes. In particular, the accuracy of telehealth preoperative assessment remains unclear. Therefore, we sought to study the prevalence of errors in diagnosis and procedure cancellations between preoperative in-person evaluations and telehealth evaluations.

Methods

We conducted a single institution, retrospective chart review of perioperative medical records at a tertiary children's hospital over a 2-year period. Data included patient demographics (age, sex, county, primary language, and insurance), preoperative diagnosis, postoperative diagnosis, and surgical cancellation rates. Data were analyzed using Fisher's exact and chi-square tests. Alpha was set at 0.05.

Results

A total of 523 patients were analyzed, with 445 in-person visits and 78 telehealth visits. There were no demographic differences between the in-person and telehealth cohorts. The frequency of changes from the preoperative to the postoperative diagnosis was not significantly different between in-person preoperative visits and telehealth preoperative visits (0.99% versus 1.41%, P = 0.557). The frequency of case cancellations between the two consultation modalities was not significantly different (9.44% versus 8.97%, P = 0.899).

Conclusions

Our results demonstrate that preoperative pediatric surgical consultations held via telehealth were neither associated with a decrease in the accuracy of preoperative diagnosis, nor an increased rate of surgery cancellations, compared to those held in-person. Further study is needed to better determine the advantages, disadvantages, and limitations of telehealth in the delivery of pediatric surgical care.

Introduction

The Institute of Medicine defines telehealth as “the use of electronic information and communications technologies to provide and support healthcare when distance separates participants”.¹ Since the early 20th century, when this idea of remote healthcare services was first introduced, telehealth has evolved remarkably.² In recent years, telehealth has been adopted in various realms of surgical care.³ Not only has this virtual platform allowed for more efficient patient education and preoperative and postoperative consultations, but it does so at reduced costs for patients and without the increased risk of potential infection exposure.4, 5, 6, 7 Some authors have gone so far as to claim that telemedicine is the “new norm” for surgical consultations and communications.⁴

According to the Centers for Disease Control and Prevention, in the year 2020 alone, Coronavirus disease 2019 (COVID-19) was the third leading cause of death, accounting for nearly 360,000 deaths in the United States.⁸ Prior to the COVID-19 pandemic, access to pediatric surgical care was already limited due to geographical discrepancies in the distribution of pediatric surgeons, with the majority of specialists located in urban centers.⁹ Specifically, in the state of Hawaii, almost all pediatric surgical care (for which over 30% of outpatient pediatric surgical encounters meet the strict criteria for telehealth delivery of care) is performed on the main island of Oahu, creating significant out-of-pocket expenses for those living on neighbor islands.¹⁰ Following the rampant onset of the pandemic and government-imposed restrictions, pediatric populations, who are already at a geographical disadvantage, became especially susceptible to gaps in quality of care.¹¹ Consequently, a telehealth revolution was born that instigated an upheaval in the usual processes by which pediatric surgical care is provided in the United States.¹¹ This urgent need for providers to rapidly change their practices allotted little time to study the safety and efficacy of these changes. As the reliance on telehealth grew, the need for large-scale research aimed at highlighting the objective benefits and possible disadvantages expanded as well.

Furthermore, while previous studies have examined the accuracy of postoperative telehealth consultations or focused on patient and provider satisfaction rates,^{4 , 12} there are only a few studies that have evaluated the diagnostic accuracy of telehealth in preoperative consultations.13, 14, 15 Therefore, the purpose of this study is to retrospectively compare the safety and efficacy of preoperative pediatric surgical consultations performed via telehealth to those performed in-person. In this way, this study will serve as a pilot study focusing on the initial implementation of telehealth in same day surgical cases to eventually aid in the planning of a larger, sufficiently powered clinical trial study of telehealth use in more complex cases and across pediatric surgical subspecialties.

Materials and Methods

Study population

We conducted a single institution retrospective chart review of perioperative medical records at a tertiary children's hospital over an approximate 2-year period from April 2019 to June 2021. While there were neither official quotas nor limitations in place at the hospital from which the data were gathered, telehealth visits were highly encouraged during late March 2020 for all patients, as deemed appropriate by the surgeon. As such, the timeframe chosen for the present study aimed to bridge any changes brought on by the COVID-19 pandemic.

This study included 523 patients. Our electronic medical record was queried for all outpatient surgical procedures for pediatric patients (age <18). Chart review was then used to identify patients who underwent elective surgery and had a preoperative outpatient visit (in-person or telehealth). All telehealth visits were carried out through synchronous high-definition video teleconferencing. All procedures were conducted by pediatric general surgeons, and the most common procedures included hernia repair, excision cyst, subcutaneous insertion port, dilation esophagus, and incision and drainage of abscess, cyst, and wound (Fig. ).

Fig.

Most common procedures by consultation modality (in-person or telehealth). The light gray represents the number of patients in the in-person cohort, and the dark grey represents the number of patients in the telehealth cohort.

Preoperative characteristics

Patient demographics included age, sex, race or ethnicity, insurance, primary language, and county. Other surgical variables included preoperative and postoperative diagnosis, procedure type, and surgical cancellation rates.

Patient stratification

While diversity of race, primary language, county, and insurance exist in this site's patient population, due to the limited sample size, patients were categorized. Races were grouped into the following categories: White/Caucasian, Black/African American, Asian (Korean, Filipino, Japanese, Chinese, Vietnamese, other Asian), Native Hawaiian (Native Hawaiian and Part Native Hawaiian), Pacific Islander (Samoan, Marshallese, Tongan, and other Micronesian), Hispanic (Hispanic or Latino, Hispanic Asian, Puerto Rican, Hispanic Native Hawaiian/other Pacific Islander), and unknown. Language was categorized into English and nonEnglish (Chuukese, Mandarin, Marshallese, Visayan, Tongan, Japanese, Cantonese, Ilocano, Kosraean, Pohnpeian, Vietnamese, and Mortlockeese). County was categorized into Honolulu, Hawaii, Kauai, Maui, and out-of-state. Insurance was categorized into commercial, Hawaii Medical Service Association, Quest (Medicaid), Tricare, and others.

Surgical outcomes

The two outcomes assessed in this study to evaluate efficacy were based on preoperative and postoperative diagnoses and cancellation rates. Differences in preoperative and postoperative diagnoses within the telehealth and in-person cohorts were compared in order to assess diagnostic accuracy, entailing the ability to evaluate the general preparedness of a patient for surgery as well as perform appropriate preoperative counseling. Frequencies of surgical cancellations were also compared.

Statistical analysis

Patients were stratified into two groups based on the type of preoperative visit: in-person and telehealth. Patient demographic and clinical characteristics were summarized by frequency and percentage for categorical variables and median with lower and upper quartiles for continuous variables. The Wilcoxon two-sample test was used to compare median patient age between the in-person and telehealth groups after assessing normality by normal probability plot as well as Shapiro–Wilk test. Fisher's exact test or chi square tests were used to compare sex, primary language, county, and insurance between the in-person and telehealth group. The chi-square test was also used to compare the number of different preoperative and postoperative diagnoses and cancellations between in-person and telehealth. All the data analyses were conducted using Statistical Analysis System 9.4, and the significance level was set at 0.05.

This study was conducted in compliance with ethical standards and reviewed by our institutional review board. It was deemed exempt due to the use of deidentified data. Additionally, an approved waiver of informed consent was obtained.

Results

Patient demographics

Of 523 pediatric patients, 445 had a preoperative visit in-person, and 78 had a preoperative visit via telehealth. Demographics between the in-person and telehealth cohorts were nonsignificant (Table 1 ). The median patient age for the in-person and telehealth cohorts were 2.00 [0.32, 11.0] and 2.50 [1.14, 9.00] (P = 0.117). Most patients in the in-person and telehealth cohorts were male (56.2% versus 56.4%, P = 0.970), English speaking (91.2 versus 97.4%, P = 0.060), and utilized Quest Medicaid insurance (53.6% versus 49.3%, P = 0.375). Unique to our study institution, most patients were Asian (38.2% versus 35.9%, P = 0.81) and native Hawaiian/Part Native Hawaiian (21.6% versus 30.8%, P = 0.06).

Table 1.

Patient demographics by consultation modality.

Demographics	In-person patients	Telehealth patients	P-value
	n (%)	n (%)
Gender
Male	250 (56.18%)	44 (56.41%)	0.97
Female	195 (43.82%)	34 (43.59%)
Age: Median [Q1, Q3]	2.00 [0.32, 11.00]	2.50 [1.14, 9.00]	0.12
Race/ethnicity
White/Caucasian	85 (19.54%)	9 (11.54%)	0.11
Black/African American	10 (2.30%)	3 (3.85%)	0.42
Asian (Korean, Filipino, Japanese, Chinese, Vietnamese, other Asian)	166 (38.16%)	28 (35.90%)	0.81
Native Hawaiian (native Hawaiian and part native Hawaiian)	94 (21.61%)	24 (30.77%)	0.06
Pacific islander (Samoan, Marshallese, Tongan, and other Micronesian)	37 (8.51%)	1 (1.28%)	0.03
Hispanic (Hispanic or Latino, Hispanic Asian, Puerto Rican, Hispanic Native Hawaiian/other Pacific Islander)	28 (6.44%)	6 (7.69%)	0.65
Unknown	15 (3.45%)	7 (8.97%)	0.03
Language
English	406 (91.24%)	76 (97.44%)	0.06
NonEnglish	39 (8.76%)	2 (2.56%)
Insurance
Hawaii Medical Service Association	125 (29.00%)	26 (35.62%)	0.25
Commercial	34 (7.89%)	8 (10.96%)	0.38
Quest	231 (53.60%)	36 (49.62%)	0.49
Tricare	34 (7.89%)	3 (4.11%)	0.25
Other	7 (1.62%)	0 (0%)	0.61
County
Hawaii	44 (10.05%)	17 (21.79%)	0.003
Honolulu	310 (70.78%)	43 (55.13%)	0.006
Kauai	20 (4.57%)	6 (7.69%)	0.25
Maui	40 (9.13%)	8 (10.26%)	0.75
Out-of-state	24 (5.48%)	4 (5.13%)	0.99

Wilcoxon two-sample test was used to compare median patient age between the in-person and telehealth group after assessing normality by normal probability plot as well as Shapiro–Wilk test. Fisher's exact test or Chi square tests were used to compare sex, primary language, county and insurance between the in-person and telehealth group.

The most common procedures utilizing telehealth were hernia repair, followed by esophageal dilation and cyst excision (Fig.). Most patients from Honolulu had preoperative visits in-person rather than telehealth (70.8% versus 55.1%, P < 0.006). Compared to Honolulu county, Hawaii county—the furthest neighboring island from the study site in Honolulu—was more likely to utilize telehealth (21.8% versus 10.1%, P = 0.003).

Outcomes

In 474 patients, the postoperative diagnoses changed in four patients (0.99%) in the in-person group and one patient (1.41%) in the telehealth group (P = 0.557, Table 2 ). The preoperative and postoperative diagnoses of the four cases in the in-person group include: incarcerated right hernia, which instead was found to be torsion of a nonviable right inguinal testis; umbilical granuloma, which instead was urachal remnant; unspecified abdominal pain which was internal hernia around the gastrostomy tube; and bleeding of unknown source, which postoperatively was declared to be no significant bleeding. The preoperative and postoperative diagnoses of the one case in the telehealth group were a case of hemorrhoids or anal polyp, which—after examination under general anesthesia—were declared to show no evidence of hemorrhoids or anal polyp. In the in-person group, 42 patients canceled their surgical cases (9.44%), and in the telehealth group, seven patients canceled (8.97%, P = 0.897).

Table 2.

Diagnosis changes and case cancellations by consultation modality.

Outcomes of interest	In-person n (%)	Telehealth n (%)	P-value
Diagnosis changes	4 (0.99%)	1 (1.41%)	0.557
Case cancellations	42 (9.44%)	7 (8.97%)	0.897

Chi-square test was also used to compare the number of diagnosis changes and case cancellations between in-person and telemedicine.

Discussion

Telehealth has been in existence for decades, yet the rise of COVID-19 has made it more applicable than ever before. The implementation of telehealth has afforded widespread access to care, high degrees of patient and provider satisfaction, and reduced costs associated with traditional in-person office visits.¹⁶ There are increasing reports of telehealth being utilized in various surgical specialties.

A prospective study in plastic surgery compared the accuracy of postoperative photos sent in by patients for the evaluation of wound healing to in-person postoperative visits, and found that there was no difference in the quality of care between the two methods.¹⁷ Similarly, a retrospective pediatric urology study evaluated the rates of postoperative surgical complications in patients followed up by telehealth compared to those who had postoperative follow-up visits in-person. The authors found there to be no postoperative surgical complications in either patient group. Of note, the authors report that the telehealth group had a significantly increased potential travel distance compared to that of the in-person group.¹⁸

While prior literature has established the advantages of postoperative telehealth visits, the role of telehealth in preoperative consultations, during which the physician is able to evaluate possible surgical candidates, overall remains to be studied further.⁴ A valid concern with respect to telehealth during the preoperative stages is the question of accuracy of the clinical evaluation without the ability to make physical contact with patients. This limitation did not prove problematic in the present study, as the telehealth cohort was neither associated with a decrease in the preoperative accuracy of diagnosis nor an increase in the frequency of surgical case cancellations in comparison to the in-person cohort among patients with similar demographics (Table 2). In the few studies that have evaluated the efficacy of telehealth in the preoperative setting, there have been reports of 100% accuracy. As early as 1998, Robie et al. were the first to report the use of telehealth in the setting of the intensive care nursery.¹⁹ Their study included six video teleconferences and six store-and-forward consultations, in which 100% of cases in both groups proved to have diagnostic accuracy.¹⁹ Similarly, in one review of 21 pediatric surgeries performed for inhabitants of remote areas of Saskatchewan, the authors reported no errors in diagnosis or changes in planned procedures.²⁰ More recently, a study by Metzger et al. explored the utility of telehealth across a diverse range of pediatric surgical subspecialties during a 2-week period in May 2020 and found no discordance in the preoperative and postoperative diagnoses for more than 98% of patients who were evaluated preoperatively by telehealth.¹⁴ Comparatively, Rizvi et al. evaluated the ability to accurately diagnose craniosynostosis or deformational plagiocephaly via telehealth and observed that 100% of patients who were diagnosed with craniosynostosis via telehealth were referred for a confirmatory computed tomography scan and underwent surgical intervention.²¹

The use of telehealth in surgical care has direct implications for the ease and accessibility of the clinical care provided—a benefit of notable importance to patients in rural communities and urban areas with limited access to medicine and transportation means.²² The unique geography of the Hawaiian Islands poses inherent challenges to healthcare delivery. According to the State of Hawaii Primary Care Needs Assessment, the neighbor islands of Hawaii are considered medically underserved areas/populations compared to the island of Oahu.²³ As all neighbor islands are considered rural, there are notable physician shortages and subsequent disparities in access to healthcare, with most of the state's population and physicians living on the island of Oahu.²⁴ Looking at physician demand among the islands, the counties of Kauai and Maui have a 3% greater demand for physicians compared to the city and county of Honolulu, and the county of Hawai'i has a 6% greater demand for physicians compared to that of Honolulu.²⁵ Importantly, the Native Hawaiian and Pacific Islander populations represent higher percentages of the local populations on the islands of Maui and Hawaii compared to Oahu, worsening the healthcare disparities in this historically-marginalized population.²⁶

For patients living on neighbor islands, where pediatric surgical providers are unavailable, they must fly to Oahu for such treatments—a harsh reality that often places a heavy financial burden on patients and their families. We found all the neighbor island counties to collectively have significantly more telehealth visits compared to Honolulu (Table 1). In attempting to gain perspective on the out-of-pocket expenses for the families participating in telehealth, we looked at the travel distance between the study institution on Oahu and the counties of Hawaii, Kauai, and Maui. The two-way travel distance ranges vary from 236 miles (Maui county) to 400 miles (Hawaii county). Our results showing that patients from the county of Hawaii were significantly more likely to use telehealth possibly reflect the increased travel distance for these patients compared to their counterparts. In addition to the burden of traveling to the clinic visit, further costly expenses worth consideration include vehicle transportation, meals, babysitting charges, and/or lodging fees. Because of the significant cost and time commitment associated with a single medical visit, profoundly more patients may be able to seek evaluation for surgical candidacy in a more time and cost-effective manner if surgeons continue to implement preoperative telehealth visits within their practices.

While these early reports and projections of telehealth are promising, there are certain ethical dilemmas associated with the use of this online modality worth consideration. In the United States alone, 4419 data breaches were reported for 314,063,186 patient healthcare records by entities covered by the Health Insurance Portability and Accountability Act between 2009 and 2021.²⁷ The growing susceptibility to data breaches makes the concern for privacy during telehealth visits a legitimate concern. As telehealth applications continue to evolve, the ability to protect patient health information should remain of high importance.

While the findings from our study are encouraging, one of the major limitations is the fact that the study is underpowered due to the small and unequal sample sizes. A posthoc power analysis revealed a power of less than 10% to identify differences in diagnosis changes and case cancellations between the two groups using the current data. As we targeted available data during the pandemic, there were not enough patient records to reach a certain power level. The primary intention of this study was to reveal the pattern and tendency of the differences as a pilot study and use our findings to aid in the future planning of a larger, sufficiently powered clinical study. Furthermore, the language barrier may have impacted the distribution of patients within each cohort, as 9% of the patients with in-person visits were non-English speakers, yet only 2.5% of the patients with telehealth visits were non-English speakers. As the option for three-way interpreter visit via telehealth was not yet established, there may have been a potential bias towards in-person visits if the patient was a non-English speaker. Similar bias may have been in effect for those who may not have felt comfortable with teleconferencing and instead preferred an in-person visit. Lastly, the retrospective design of the study is also a limitation, as the data analysis may have been privy to missing data. For example, not all cancellations were associated with a documented reason. Thus, all cases that were canceled—whether patient or family initiated due to concern for COVID-19 exposure, for example, or surgeon initiated due to no true indication for surgery on the day of arrival—were treated equally.

Future work will be aimed at broadening the scope of the current study to evaluate the efficacy of telehealth in other pediatric surgical subspecialties, to include pediatric orthopedic surgery, urology, otolaryngology, and neurosurgery. We anticipate similar efficacy rates of telehealth in these surgical subspecialties to what was observed within the current study, which would afford great value to an expansive patient population. With this broadened inclusion criteria, different subspecialty considerations would then come into play—for example, a heavier reliance on other elements of the electronic health record, such as imaging studies in orthopedic surgery. Additionally, future efforts will focus on assessing telehealth efficacy in more complex diseases. As this study specifically looked at same-day surgery cases (which by nature are more straightforward, less complicated cases), we are interested in discerning the efficacy rates of telehealth when used for more complex cases as the use of telehealth at our institution continues. Furthermore, as reimbursement for teleconferencing can act as a barrier to large-scale implementation, the authors intend to evaluate reimbursement patterns from different insurance payers to better elucidate the feasibility, efficacy, and utility of telehealth.

Conclusions

In conclusion, the favorable clinical outcomes observed with the use of telehealth in the preoperative setting with regards to diagnostic accuracy and case cancellation rates underscore the benefits of its use and the very real opportunities for gross expansion.

Author Contributions

The authors confirm their contributions to the paper as follows: study conception and design: Russell Woo, Benjamin Tabak, Ashley Shirriff; data collection: Ashley Shirriff, Anna Gragas; analysis and interpretation of results: Hyeong Jun Ahn, Maveric Abella; draft manuscript preparation: Ashley Shirriff, Anna Gragas, Maveric Abella, Hyeong Jun Ahn, Benjamin Tabak, and Russell Woo. All authors reviewed the results and approved the final version of the manuscript.

Disclosure

Dr Russell Woo is an Associate Editor for the Journal of Surgical Research; as such, he was excluded from the entire peer review and editorial process for this manuscript.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.