Compliance of Tonsillitis e‐Health Patients: A Prospective Cohort Study

ABSTRACT

Objective

As canceled appointments and undertreated patients generate substantial expenses in healthcare, this study examines whether tonsillectomy candidates assessed via telemedicine‐assisted care paths exhibit elevated appointment cancellation rates compared to those evaluated in‐person at an outpatient clinic.

Methods

At Helsinki University Hospital, a tertiary referral clinic, two telehealth‐assisted care paths (virtual visits and a digital care pathway) have been available in recent years to remotely assess patients for tonsillectomy. To evaluate whether preoperative telehealth negatively affects tonsillitis patients' compliance to treatment, we compared cancellation rates of these two e‐health care paths as primary outcome measures with the outpatient clinic, which was defined as the control group.

Results

The remote care paths showed the highest rates of cancellation. 25.9% of the digital care pathway patients canceled their care path compared to 6.6% of outpatient clinic patients who canceled appointments (p < 0.01). 29.4% of the digital care pathway patients had an additional remote contact with the clinic preoperatively compared to 18.5% of patients assessed at the outpatient clinic (p = 0.002). Patients assessed remotely were more likely to make an unplanned additional contact than patients seen at the outpatient clinic. No significant differences in canceled surgeries were observed between the cohorts.

Conclusion

Assessing patients for tonsillectomy via a digital care pathway appears to increase cancellations and additional contacts. Therefore, both digital and in‐person assessments should remain available for tonsillectomy candidates to accommodate varying patient needs and offset additional costs.

Level of Evidence

Level 3.

As canceled appointments and undertreated patients generate substantial expenses in health care, this study examined whether tonsillectomy candidates assessed remotely for surgery exhibited higher cancellation rates. Our findings indicate that patients assessed via telehealth were more likely to interrupt their care path, leading to an increased need for additional contacts compared to those evaluated in‐person. Based on these results, we recommend that both digital and in‐person options are offered to tonsillectomy candidates to accommodate diverse patient needs.

1. Introduction

The use of telemedicine has increased exponentially in the last few years, largely due to the COVID‐19 pandemic and the need for new treatment options [1]. While telemedicine has been shown to improve access to healthcare overall, some studies have shown that it may increase no‐shows for certain demographics and medical conditions [2]. Research on patient compliance with e‐health treatment is preliminary, and more research is needed to evaluate its benefits. Many telemedicine studies report high patient satisfaction rates but lack the use of validated questionnaires to assess patient compliance [3]. Nevertheless, telemonitoring has been found to improve adherence to CPAP therapy in some studies, and systematic reviews of e‐health interventions have reported improved medication adherence among patients with type 2 diabetes, as well as improved treatment adherence and outcomes in those with chronic diseases [4, 5, 6]. Telemedicine may also support patients with affective disorders or severe mental illness adhere to treatment plans [7, 8]. However, negative experiences with online disease‐related searches may adversely affect patients' adherence to medication use [9].

Tonsillectomy (TE) is one of the most frequently performed operations in ear, nose, and throat (ENT) practice [10]. Patients evaluated for TE using a digital care pathway (DCP) suffer no more postoperative complications or follow‐up visits than those evaluated at the outpatient clinic (OPC) [11]. Significant savings in cost and time were also achieved in the remote assessment of these patients [12].

Research on patient compliance in the treatment of tonsillitis or in the field of ENT is limited. Given that canceled appointments and undertreated patients generate substantial expenses in healthcare, this study examines whether TE candidates treated via telemedicine‐assisted care paths cancel more appointments than those evaluated in person at the OPC, potentially indicating a weaker commitment to their care.

2. Materials and Methods

2.1. Study Design and Participants

This prospective cohort study included 800 TE candidates referred to Helsinki University Hospital Department of Otorhinolaryngology—Head and Neck Surgery, the only public tertiary care unit performing TEs within a catchment area of 1.6 million inhabitants. Patients under 17 years of age or not treated according to the study protocol were excluded. Of the 800 patients, 255 were directed to the DCP, 197 to a virtual visit (VV), and the remaining 348 to the OPC for preoperative assessment. Patient episodes occurred between August 2020 and November 2022, March 2020 and June 2022, and March 2021 and March 2023 for OPC, VV, and DCP, respectively. The OPC cohort was defined as the control group to investigate the impact of telehealth. Patient allocation was based on the clinic's existing criteria, and no randomization was applied: Patients were assigned to the OPC if any exclusion criteria for VV or DCP were present in the referral (Figure 1). Patients eligible for remote assessment for TE typically had a history of recurrent or chronic tonsillitis or peritonsillar abscesses, were aged between 18 and 39 years, and were suitable for day surgery [13].

FIGURE 1.

Exclusion criteria for remote preoperative assessment for tonsillectomy.

2.2. Methods

At the tertiary referral clinic, three preoperative care paths for TE candidates have been available over previous years: OPC, VV model, and the DCP. To evaluate the impact of telehealth on cancellation rates, the OPC cohort served as the control. The primary outcome measure was the cancellation rate of various appointments. Secondary measures included the proportion of operated patients and the lead time of each care path.

The conventional OPC model included one preoperative visit to the clinic during which an ENT specialist assessed the patient for TE or alternative lines of treatment, such as watchful waiting (Figure 2). In the VV model, patients received a patient information and questionnaire package via email, prior to a scheduled VV via a phone call with an ENT specialist, which replaced the in‐person appointment. The DCP replaced the preoperative assessment appointment with an interactive online platform containing videos, written information, and questionnaires regarding the patient's tonsillitis history and general health (Figure 3). Having completed the DCP, patients answered online whether they were interested in undergoing TE based on the received information, or if they wished to consult with a doctor before deciding. An ENT specialist reviewed the patient's answers and either (1) assigned the patient to TE, (2) contacted the patient by phone for further information, or (3) referred the patient to the OPC for a clinical examination. If satisfied with the information on the DCP and in the absence of emerging medical reasons, DCP patients had no direct contact with an ENT specialist before the day of surgery. The surgery date was scheduled approximately 1 month prior to the operation. On the day of surgery, the operating surgeon confirmed the validity of the operation's indication and checked for contraindications. In each preoperative care path, the eligibility for TE required both the fulfillment of the international guidelines and the willingness to undergo the procedure [14].

FIGURE 2.

Planned process flows of tonsillectomy candidates.

FIGURE 3.

User experience of the digital care pathway for tonsillectomy evaluations.

Patients were prospectively collected with a 30‐day postoperative follow‐up. Information on diagnoses, canceled appointments, additional and unplanned contacts, rescheduled and canceled surgeries, lead times, and demographic data were manually extracted from the electronic patient record system.

2.3. Statistical Analysis and Variables

Cancellation and rescheduling rates were calculated as the proportion of patients who canceled or rescheduled appointments out of the total patient cohort. A canceled DCP was defined as an unfinished DCP, regardless of the reason. Cancellations of VV or OPC appointments included unplanned rescheduling and no‐shows. All cancellations resulted in a termination of care at the clinic, meaning that these patients had to obtain a new referral from primary care for reassessment or seek private‐clinic treatment at their own expense.

Additional contacts were defined as planned interactions that were not initially scheduled within the planned process flows. Unplanned contacts included patient‐initiated interactions with the clinic without a scheduled follow‐up or prior notice. Contact methods comprised messages, calls, and emergency department visits. All contacts with a patient record entry were included. Preoperative appointments included all contacts prior to surgery. Rescheduled surgeries were identified in the patient record system through communications with the nurse in charge of the OR schedule. Canceled surgery was defined as refusing the operation despite being initially eligible for it. Lead time was measured as the number of days from receiving the referral to the end of the 30‐day postoperative follow‐up.

To investigate differences between care paths, 2 × 2 contingency tables were created for categorical variables, including cancellations, additional and unplanned contacts, rescheduled and canceled surgeries, and eligible and operated patients. χ ² analysis with odds ratios (OR) was applied for all contingency tables. t‐Test was performed on continuous variables such as lead time.

A significance level of p < 0.05 was considered statistically significant. Calculations were performed using Microsoft Excel 16.77.1 (Microsoft Corp., Redmond, WA, USA) and IBM SPSS Statistics 29.0.0.0 (241) (IBM Corp., Armonk, NY, USA).

2.4. Ethical Approval

The study was approved by the Ethics Committee of Helsinki University Hospital, Helsinki, Finland (§200 HUS/2242/2020; §130 HUS/1895/2021) and research permission was granted by the Research Administration of the hospital (§105 HUS/85/2021; §69 HUS/146/2020). This prospective cohort study was conducted in accordance with the Declaration of Helsinki [15].

3. Results

Patient episodes of 800 TE candidates across three separate preoperative care paths were analyzed. Most patients were female (581/800, 72.6%) (Table 1). 87.4% of the patients were between 18 and 39 years old.

TABLE 1.

Patient demographics of 800 tonsillectomy candidates.

	Male (n)	Female (n)	Patients (n)	% of all patients	Digital care pathway (n)	Virtual visit (n)	Outpatient clinic (n)
Age group (years)
17–19	23	70	93	11.6	31	23	39
20–29	106	322	428	53.5	166	123	139
30–39	58	129	187	23.4	56	46	85
40–49	18	33	51	6.4	1	4	46
50–59	11	20	31	3.9	0	1	30
60–69	2	5	7	0.9	1	0	6
70 or over	1	2	3	0.4	0	0	3
Total	219	581	800		255	197	348

3.1. Cancellations of Preoperative Appointments

The remote care paths showed the highest cancellation rates. 25.9% of the DCP patients canceled the DCP compared to a 6.6% OPC cancellation rate (p < 0.01) (Table 2). 12.2% of VV patients canceled their VV, showing a significant difference compared to the DCP cohort (p = 0.02). No statistically significant differences in cancellation rates were discovered between the VV and OPC cohorts (p = 0.05).

TABLE 2.

Preoperative cancellations and additional contacts.

	Digital care pathway (DCP)	Virtual visit (VV)	Outpatient clinic (OPC)	OR (95% CI), DCP vs. OPC	OR (95% CI), VV vs. OPC
Cancellation	66/255 (25.9%)	15/123 (12.2%)	23/348 (6.6%)	4.93 (2.97–8.20)	1.96 (0.99–3.90)
Additional outpatient clinic appointments	32/243 (13.2%)	6/187 (3.2%)	N/A	N/A	N/A
Additional call/virtual visit	75/255 (29.4%)	11/187 (5.9%)	63/341 (18.5%)	1.84 (1.25–2.70)	0.28 (0.14–0.54)
Unplanned visit/call/virtual visit	32/255 (12.5%)	38/197 (19.3%)	38/345 (11.0%)	1.16 (0.70–1.91)	1.93 (1.18–3.15)

Note: Statistically significant findings (p < 0.05) in bold, reference group: outpatient clinic.

3.2. Additional Appointments and Contacts

13.2% of the DCP patients had an additional in‐person visit, compared to 3.2% VV patients (p < 0.001). These additional visits were mostly consequences of patients not proceeding on the DCP or needing further clinical examination. Patients with chronic tonsillitis, who were predominantly referred to the DCP, appeared overrepresented in additional in‐person visits. While 35.7% of patients on the DCP had chronic tonsillitis, these patients accounted for 61.3% of the additional in‐person visits (p = 0.001).

Preoperative additional remote contacts were higher among DCP patients (29.4%) compared to OPC patients (18.5%) (p = 0.002). However, VV patients contacted the clinic remotely significantly less often than the OPC patients, at only 5.9% (p < 0.001).

Remotely assessed patients were more likely to have an unplanned contact than OPC patients. 19.3% of VV made an unplanned contact compared to 11.0% of OPC patients (p = 0.008). Between the DCP and the OPC, no significant difference was discovered. Unplanned contacts were mostly calls to the clinic regarding the surgery and OPC appointments regarding emerging symptoms. All additional and unplanned visits included, 139 additional visits were identified among the 255 DCP patients (54.5%), 55 for 197 VV patients (27.9%), and 101 for 345 OPC patients (29.3%).

3.3. Rescheduled and Canceled Surgeries

Among the cohorts, 66.7% of the DCP patients, 82.7% of VV patients, and 46.0% of OPC patients met the criteria for TE and were willing to undergo surgery (Table 3). Fewer OPC patients were eligible for surgery compared to DCP (p < 0.001) and VV (p < 0.001) patients.

TABLE 3.

Surgical eligibility, frequency, and cancellations of tonsillectomy patients across three preoperative care pathways.

	Eligible for surgery	Operated patients	Rescheduled surgeries	Canceled surgeries
Digital care pathway	170/255 (66.7%)	141/255 (55.3%)	63/170 (37.1%)	29/170 (17.1%)
Virtual visit	163/197 (82.7%)	140/197 (71.1%)	51/163 (31.3%)	23/163 (14.1%)
Outpatient clinic	160/348 (46.0%)	141/348 (40.5%)	44/160 (27.5%)	19/160 (11.9%)

Note: Statistically significant findings (p < 0.05) in bold, reference group: outpatient clinic.

DCP patients were the most likely to reschedule the surgery at least once. However, no significant differences in cancellation rates were found between cohorts (DCP/OPC p = 0.06; VV/OPC p = 0.45; DCP/VV p = 0.27).

Of all DCP patients, 55.3% underwent TE, whereas 71.1% of VV patients were operated on (p < 0.001). OPC patients were the least likely to be operated on, at 40.5%. DCP patients were the most likely to refuse the surgery despite being eligible for it. However, no significant differences in canceled surgeries were discovered between the groups.

The median lead time for operated patients across all cohorts was 255 days. Median lead times were 264 days for DCP, 259 days for VV, and 236 days for OPC, with no significant differences between cohorts.

4. Discussion

This prospective study compared the cancellation rates of 800 TE candidates treated via three different preoperative care paths: conventional in‐person visits at an OPC, phone call‐based VVs, and a DCP. Being one of the most common surgical procedures, TE causes substantial disease and economic burden on both the patients and healthcare providers [10]. Therefore, cost‐effective, medically justified, and appropriate treatment paths for both patients and healthcare providers are needed. The DCP in this study has demonstrated cost savings compared to the traditional OPC model, whereas the VV model reduced patient fees but did not lead to savings for the healthcare provider [12]. Furthermore, postoperative recovery was similar in patients assessed preoperatively by DCP as in patients evaluated in‐person [11]. This study investigated the applicability and practicality of the DCP and VVs from the perspective of the patient and the healthcare system.

The completion rate of the DCP (74.1%) was similar to that of other digital care programs [16]. However, the cancellation rate was significantly higher for DCP compared to both VV and OPC, which aligns with scarce literature suggesting that certain telemedicine applications may adversely impact no‐show rates in specific patient groups [2]. Unlike the OPC, which required only an in‐person clinic visit, and the VV, where patients received a pre‐call information package but were not mandated to review it, the DCP required that patients actively engage with the digital platform, read materials, fill out questionnaires, and watch informational videos. Thus, the DCP required more patient involvement, which may largely explain the lower completion rate. However, patients spent less time completing the DCP (62 min) compared to an OPC visit, traveling included (2–4 h) [12]. Technical issues with the DCP platform could also contribute to the lower completion rate. Although a survey conducted at our clinic found that most patients experienced the DCP platform as user‐friendly and reliable, a systematic qualitative study could provide insight into the lower completion rates. Additionally, the OPC cohort differed slightly; for example, patients presenting with more concerning symptoms were directed to the OPC, perhaps making them more willing to complete treatment than patients in DCP and VV. Care path cancellations may also indicate mild symptoms, and therefore potentially low benefit of TE. However, since TE is considered a quality of life improving treatment, undertreatment may increase absence from work and morbidity [17].

A higher amount of additional in‐person visits (13.2%) and remote contacts (29.4%) in DCP patients also seems partly due to individual challenges in progressing on the DCP. On the other hand, VV patients had considerably fewer additional face‐to‐face visits (3.2%) and remote contacts (5.9%). This difference may arise from the different patient activity requirements of these two telehealth paths but also from the real‐time communication between patients and healthcare providers during the VV, allowing patients to ask questions and clarify uncertainties, reducing the need for further in‐person contact [18]. Additionally, patients with chronic tonsillitis appeared overrepresented in the group requiring additional visits. This may suggest that the decision to proceed with surgery is particularly difficult for these patients, potentially reducing their suitability for telemedicine‐based assessments.

The reason why significantly more patients in telehealth paths were eligible for surgery than patients in OPC may reflect differences in inclusion criteria of the paths. Patients with higher surgical and/or anesthetic risk (e.g., high BMI and older age) were referred to the OPC, where the decision to operate was less frequent, probably due to these risks. It can also be speculated that some patients may trust the doctor more than written information on surgery risks, which could reduce the number of OPC patients referred to surgery.

Due to medical reasons and patient safety, randomization was not feasible for this patient population. Patients presenting with potentially serious symptoms or underlying conditions were assigned to the OPC path. Additionally, the inability to randomize was influenced by hospital policy, as VVs were discontinued with the implementation of the more cost‐efficient and patient safe DCP [11, 12]. However, there were no differences in the inclusion criteria between VV and DCP groups. Nonetheless, all patients were treated according to the clinic standards, and potential biases have been discussed in detail.

In each care path, more than a quarter of the patients rescheduled their surgery date at least once, with no statistically significant differences between the groups. However, the overall rescheduling rate (27.5%–37.1%) can be considered high, causing increased work and costs for the healthcare provider. In addition to reduced compliance, rescheduling may have reflected milder symptoms, young adults' life priorities, and hesitancy regarding the surgery and the recovery time. Lengthy lead times, from referral to the end of the 30‐day follow‐up, have been associated with higher cancellation rates in some studies [19]. Nevertheless, Finns are accustomed to long waiting times in the public sector, particularly after COVID‐19, which further extended delays. Future studies should analyze the reasons behind rescheduling in more detail to improve the efficacy of digital care paths. Incorporating a modest patient fee for DCP would potentially lower cancellation rates.

While this study was conducted in the context of Finnish healthcare, TE management processes are highly similar across healthcare systems. Therefore, the DCP could be applied in more fragmented healthcare systems, such as in the United States, where it could function as a self‐referral or screening tool prior to an in‐person visit. By providing standardized patient information, the DCP has the potential to equalize patient treatment and lower barriers to seeking care, particularly for those in rural areas, irrespective of the healthcare system. Furthermore, the DCP has potential applications in low‐income countries, where it could streamline consultant services, which are often limited.

Digital platforms can improve treatment and patient equity by providing all patients with standardized information about the benefits and disadvantages of TE versus conservative treatment [2, 20]. However, it is important to understand the diverse needs and the expectations of different patient groups [2]. Although patients are increasingly accustomed to telemedicine as the population ages, previous studies have shown that older patients are less likely to use digital care services than younger patients [21]. Most tonsillitis patients are adolescents and young adults, who are accustomed to digital solutions and may find a DCP more convenient than visiting an OPC. Nevertheless, healthcare providers should continuously modify and improve their digital care platforms according to patient feedback. Based on this study, we recommend that both digital and in‐person options are offered to TE candidates to accommodate diverse patient needs. The goal of digitalization is to increase the availability of treatment, improve and equalize the quality of treatment, and reduce costs without excluding some groups of patients.

In summary, our previous studies support the DCP as a feasible, safe, and cost‐effective method in the preoperative assessment of TE candidates. However, the results of this study emphasize the need for comprehensive telemedicine evaluations using standardized healthcare management frameworks, such as the Triple Aim, which ensures both effectiveness and patient experience across diverse clinical settings [22].

5. Conclusion

Patients on the DCP exhibited higher cancellation rates, leading to an increased need for additional preoperative contacts compared to those assessed in‐person. Based on these findings, it is recommended that TE candidates are offered both digital and in‐person options to accommodate varying patient needs. Nonetheless, our previous research along with this study demonstrates the DCP as a feasible, medically justified, and cost‐effective method for preoperative assessment of TE candidates.

Conflicts of Interest

The authors declare no conflicts of interest.

Hakanen O., Lamminmäki S., Torkki P., and Tolvi M., “Compliance of Tonsillitis e‐Health Patients: A Prospective Cohort Study,” Laryngoscope Investigative Otolaryngology 10, no. 3 (2025): e70184, 10.1002/lio2.70184.