Knowledge of Day and Time Contact Preferences Assists in Successful Telephone Follow-Up

Telephone call follow-up in clinical trials has become increasingly important, especially in the era of the COVID-19 pandemic due to the restrictions on in-person contact [1]. Calls serve as a viable alternative to face-to-face visits, particularly in clinical populations such as people living with stroke [2‒6]. In this population, limited mobility, financial constraints, driving restrictions, living remotely, and co-morbid conditions may impact the ability to attend study visits in person [7, 8]. Telephone call follow-up provides benefits such as reducing the burden of participation in clinical trials, leading to improved response rates and fewer dropouts [9‒11]. Often, telephone contact is preferred when compared with face-to-face in a range of clinical scenarios, such as pre- and post-operation, healthcare consultations, receiving health information, and clinical trial follow-up [12‒15]. However, telephone calls may be more costly than other means of data collection, such as postal surveys, because of the costs to employ research staff to service multiple contact attempts [16]. This emphasizes the need to improve telephone call contact methods in clinical trials. Determining up front the participants’ preferences for follow-up telephone calls may facilitate response rates and improve the chance of successfully completing follow-up assessments on time [9, 17].

Utilizing extensive contact data, collected during the ReCAPS trial (Recovery-focused Community support to Avoid readmissions and improve Participation after Stroke), we examined how to optimize telephone contact, used as the primary means to complete study assessments in participants within the first 3 months after stroke. We examined (i) participants’ telephone contact preferences, and these preferences related to the day and time of actual telephone contact made; (ii) the number of contact attempts needed to successfully complete telephone assessments; (iii) the association between participant characteristics and the number of contact attempts made and between participant characteristics and agreement of time/day of call with participant preferences.

The ReCAPS trial is a randomized control trial, which is testing a new eHealth self-management support program for people living with stroke [18]. It involves person-centered goal setting and randomization via a telephone call conducted 7–14 days after hospital discharge and subsequently sending information via text message or email for a period of 3 months. Participants are followed up via telephone call to complete assessments at 3 months post-randomization (when primary and secondary outcomes are collected) and at 12 months post-randomization. The predetermined sample size for ReCAPS is 890 participants.

Patients who were admitted to stroke wards in nine hospitals around Australia (five sites in Victoria, two in New South Wales, one in Queensland, and one in South Australia) were screened by trained clinicians for the following inclusion criteria: aged ≥18 years; confirmed diagnosis of acute stroke; discharged directly to a home setting from a stroke unit within 14 days of admission; have access to the Internet; self-identify as user of text message/email technology; able to communicate in English; a baseline-modified Rankin Score (mRS) of 0 (no symptoms) to 4 (moderately severe disability); able to consent. Patients with significant language impairments who were unable to communicate their goals and those who were discharged to in-hospital rehabilitation or with a poor prognosis (within 3 months) were excluded. Consent was obtained by clinicians either in person or using e-consent. Recruitment for ReCAPS is ongoing. On average, two participants were recruited each week from August 2019 to April 2022.

This is a prospective observational study nested within the ReCAPS trial. A pragmatic sample size was based on a minimum of 200 participants completing the 3-month follow-up assessments. Data for the analysis for this study were closed off on April 14, 2022. Study data were collected using a purposefully designed database created using REDCap (Research Electronic Data Capture) tools hosted at Monash University [19, 20]. REDCap is a secure, Web-based software platform designed to support data capture for research studies.

As part of demographic data collection at recruitment, participants were asked to specify their contact preferences for scheduled telephone call assessments. For the preferred day of the week, participants were asked to specify at least three options from Monday to Friday or specify “any weekday.” For the time of day, participants were asked to specify at least one option from “morning”, “afternoon,” and “evening (6:00 pm–8:00 pm),” or specify “any time.” The research personnel used these contact preferences to schedule contacting the participant for their assessments.

Telephone calls were made by three part-time members of the Monash University ReCAPS research team who cover Monday to Friday. Assessments took approximately 45 min to an hour to complete. Multiple telephone call attempts to contact the participant were made when needed. This involved calling as many recorded contact sources for the participant as required to complete the study assessments. A successfully completed telephone call was defined by the collection from the participant of all required data for the study assessments. Research personnel were asked to conduct assessments based on the participant’s reported contact preferences wherever practical. The day and time of each contact attempt were recorded in REDCap.

To record the time and day of a successful telephone call, a time stamp was used that is generated by REDCap when assessment data are entered by the researcher during the start of the call to the participant. Time stamps were generated according to the time in Melbourne, Australia, and were adjusted for those participants living in different time zones across Australia.

Descriptive statistics were used to summarize the proportion of contact preferences, number of contact attempts, and successful completion of assessments. Logistic regression was used to examine the association between participant characteristics (sex, age, employment status, and living alone) and (a) the number of contact attempts (within two attempts vs. 3+) and (b) successful contact on preferred day/time. Results are reported as odds ratios with corresponding 95% confidence intervals. The significance threshold was set at p value of 0.05. The data were analyzed using STATA/SE version 17.

Data from 232 participants were available for this study: average age 66 years (SD = 12), 69% male, 94% ischemic stroke. The majority of participants lived with others (80%), and 56% were not working prior to their stroke (including being retired, unemployed, or on sick/long-term leave).

Approximately 59% of participants preferred to be contacted on “any day” and 28% of participants chose three preferred days (5% chose 4 days, 3% chose 2 days, and 5% chose 1 day). Wednesday was the preferred day to be contacted (26%), followed by Monday (21%), Thursday (19%), and Tuesday (18%), with Friday being the least preferred (16%, online suppl. Table I; for all online suppl. material, see https://doi.org/10.1159/000535639). Regarding the preferred contact time, most participants (65%) chose one preferred time and 25% of participants chose “any time.” Morning was the preferred time of day (49%), followed by afternoon (32%), with evening being the least preferred (19%, online suppl. Table II).

At the time of data extraction, there were 208 participants who had reached the time point for their 3-month follow-up assessment. Only six of these (3%) did not complete their telephone assessments (four, uncontactable; two, initial contact was made and the participant requested to reschedule but was subsequently uncontactable).

Sixty percent of telephone assessments were successfully completed when calls were made on the preferred day and time nominated by the participant at baseline (Table 1). Telephone study assessments were more successful if they were made on the participant’s preferred day (even if the time was not preferred), compared to if they were made during the participant’s preferred time (if the day was not preferred). For example, 26% of assessments were successfully completed on a participant’s preferred day only, compared to 10% of successful assessments made during a participant’s preferred time only (Table 1).

Nearly half of the successful follow-up calls were completed within two contact attempts (47%) and 71% of telephone calls were successfully completed within three attempts (Table 2).

There was no association between participant characteristics and the number of contact attempts made for the telephone call assessment (online suppl. Table III). There were no characteristics associated with the successful completion of assessments on the preferred time/day for the telephone call (online suppl. Table IV, V).

We provide new evidence, based on a case study in stroke, that highlights the importance of knowing a participant’s preferred contact day and time for the efficient completion of study assessments via a telephone call. We found a higher proportion of assessments conducted via telephone calls were successfully completed on the day and time specified by the participant. In addition, nearly half of the calls were completed within two contact attempts. These findings suggest that determining a person’s preferences may facilitate the timely completion of telephone assessments.

Although the majority of participants indicated they would be happy to be contacted on any day, of the specific weekdays, Wednesday was preferred. The preferred time to be contacted was in the morning and the least preferred time to be contacted was in the evening. These findings are consistent with those of previous research conducted by Mapetla et al. [9] and Learmonth et al. [17] (respectively). Mapetla et al. [9] conducted a survey among vaccine trial participants, asking their hypothetical preferred time of day to be contacted by the researchers. These authors reported that 49% of participants preferred to be called for trial communication in the morning, while 31% preferred the afternoon. Learmonth et al. [17] conducted a RCT with 57 participants with multiple sclerosis with an intervention involving behavioral coaching calls. They also showed that evening was the least preferred time to receive calls (19%). However, in contrast to our finding of the morning being the preferred contact time, they concluded that the afternoon was preferred (55%), with 26% of participants preferring the morning. Taken together, this information on participant preferences may be used to assist with the timing of telephone or video calls in future studies. It could also potentially be useful when requiring participant responses to digital communication such as text messages or emails.

Our results showed that nearly 50% of telephone call assessments were successfully completed within two contact attempts. The contact preferences of participants were typically adhered to by the research team, but in the instances that they were not, this was primarily due to time constraints of the researchers, or the participant requesting another specific time for contact. Telephone call follow-up has been shown to be more costly due to the costs of employing research assistants to have flexible working hours and make multiple attempts to complete telephone call assessments [5]. Given this, determining a participant’s preferred contact time for trial assessments may reduce the number of unsuccessful contact attempts, with a positive impact on the cost and feasibility of telephone call-based research.

There was no significant association between employment status and telephone call assessments being successfully completed on a participant’s preferred contact day and/or time. These results may have been influenced by the strict and lengthy lockdowns that occurred during the COVID-19 pandemic in Australia. Most of the study assessments were conducted during the COVID-19 pandemic and data on the mode of working (working from home vs. in office) was not explicitly collected. This may have resulted in a greater proportion of people working from home than would normally be expected. Therefore, more participants might have been available to receive a telephone call. This may limit the generalizability of the study findings for people who do not work from home.

In conclusion, determining a participant’s contact preferences may assist in the successful completion of study assessments via telephone. Sixty percent of telephone call assessments were completed by research staff on the day and time that agreed with the participant’s preference, with an additional 30–35% completed at either the agreed time or day. In addition, study assessments were successfully completed within two or three contact attempts for 70% of participants. To optimize participant engagement and reduce research costs, we recommend that research staff be available for telephone contact on Wednesdays and in the mornings. This information may support future design of telephone-based data collection methods. Ultimately, this has the potential to have a positive impact on the feasibility of both clinical trials and longitudinal telephone assessments.

The study was approved by the Human Research Ethics Committee: Monash Health (RES-18-0000-170A); Monash University (Project ID 16435); Peninsula Health (SSA/39945/PH-2019); Eastern Health (SERP93-2018); Austin Health (SSA/39945/Austin-2019-166787); Royal Prince Alfred Hospital (SSA/2019/STE16207); St Vincent’s Hospital Sydney (SSA/2019/STE16711); Alfred Health (SSA/39945/652/19); Sunshine Coast University Hospital (SSA/2020/QSC/39945); Royal Adelaide Hospital (SSA/19/CALHN/494). Written informed consent was obtained by clinicians either in person or using e-consent.

The authors declare that they have no conflict of interest.

The ReCAPS phase III trial is funded by a project grant from the National Health and Medical Research Council (NHMRC, 1162596) from 2019 to 2023. The following authors received Research Fellowships: DAC (NHMRC 1154273), NAL (Future Leader Fellowship from the Heart Foundation 102055), MFK (Future Leader Fellowship from the Heart Foundation 105737).

O.B. and M.R. wrote the paper and performed the data analysis. All authors were involved in the conception, design, and interpretation of the study and critically reviewed the manuscript during the writing process, approving the final version to be published.

All data inquiries can be directed to the corresponding author.