Evaluation of an internet-based aftercare program to improve vocational reintegration after inpatient medical rehabilitation: study protocol for a cluster-randomized controlled trial

Patients are recruited during inpatient psychosomatic, orthopedic or cardiovascular rehabilitation. Eligible patients have to be vocationally strained, assessed by the ‘Screening Instrument zur Messung des Bedarfs an berufsbezogenen Behandlungsangeboten in der medizinischen Rehabilitation’ (Short Screening Instrument for the Assessment of Need for Occupation Related Treatment in Medical Rehabilitation; SIBAR [6]), a short screening questionnaire measuring the social medical risk of early retirement, occupational stress and the subjective need for occupational treatment. Patients are eligible if they report high occupational stress or risk of early retirement appropriate to the cutoff reported by the authors of the SIBAR or if they state a high subjective need for occupational treatment. Furthermore, eligible patients have to be German speaking, currently employed, between 18 and 59 years old and have private internet access, so that they do not have to use internet access at work for participation in our study and are not disturbed by others. Patients are excluded if they are unemployed or retired, or have severe physical or psychological complaints, precluding participation in the stress management training (GSA). We also excluded anyone aged over 59 because the likelihood of retirement is very high over 59 years, limiting therapeutic access to these patients. The Study Centre of Mental Disorders at the University Medical Centre of the Johannes Gutenberg University of Mainz is responsible for storing personal data, encoding the participants and randomizing the groups. Administration of the internet platform, allocation of the weekly writing tasks and therapeutic feedback are managed by psychologists of the Department for Psychosomatic Medicine and Psychotherapy of the University Medial Centre Mainz. The clinical staff of the four rehabilitation centers (one psychosomatic, one orthopedic and two cardiovascular) conducting the inpatient GSA program and introducing the internet-based aftercare to the patients consists of social education workers, psychologists and psychotherapists, all experienced in vocational stress management and psychological group training.

Data security is guaranteed by secure sockets layer-coded internet connections as used in bank transfers and a firewall-protected webserver for both the MySQL-database and the application of the internet platform. All patients are instructed to use pseudonyms to login and for all actions on the internet platform so that no personal data is stored on the webserver; therefore, identification of the user is not possible.

Over a period of 18 months, a total of 800 patients is planned to be included in the study (100 rehabilitants in each center for each randomized arm).

At inpatient admission, patients are informed in detail about the study by the clinical staff of the rehabilitation centers. After giving written consent and screening, the eligible patients are assembled into a group of up to 12 (minimum of two) participants. Over a period of two to three weeks, participants are trained in stress perception and management by four interactive psycho-educative, cognitive-behavioral and psychodynamic modules (90 minutes each). To standardize the procedure, the staff has been trained using a comprehensive manual in a detailed one-day seminar before study start, with a refresher half-way through recruitment. In the first part of the fourth module, the participants are familiarized with the three components of Luborsky’s Core Conflictual Relationship Theme and its recurrent and potentially mal-adaptive nature, focusing on typical expectations regarding return to work [43, 44]. This approach also serves as a basis for the upcoming writing task in the intervention group. The introduction to the features of the internet platform, given in the last of four inpatient GSA modules, differs between the two randomized arms. Up to the fourth session, participants and trainers are blind to the result of randomization. Immediately prior to the fourth module, the trainer is unblinded. In the second part of the fourth session, all participants are introduced to the special features of the internet platform. Instructions for participants depend on the arm they are randomized to (intervention versus control). Because there is plenty of interaction between consecutive cohorts of patients admitted to medical rehabilitation, we do not randomize individuals but groups of eligible patients. With this cluster randomization, we aim to minimize contamination through the interaction of patients during inpatient rehabilitation who are getting different instructions for the aftercare interventions in the two randomized arms.

In two of the four rehabilitation centers, patients of both randomized arms have the opportunity to log on to the study website during their inpatient rehabilitation treatment and complete the baseline questionnaire online. In the other two clinics without sufficient technical equipment, the participants get the same questionnaire on a paper-pencil basis and are prepared for their first login at home by the trainer. In any case, every participant gets a sealed envelope including the personal access data and a written description of the intervention or control program. After login, all patients are asked to change their password, to choose a nickname and to enter their e-mail address if they want to be informed by automated e-mails about study-related information, therapeutic tasks and questionnaires to be filled out.

The essential component of the online aftercare program for the intervention group is a standardized weekly writing task (‘blog’) followed by individual feedback from the online therapist within one week day, both only visible to the individual patient and the online therapist. Further online features for the intervention group are a self-test with computer-generated feedback to the participants’ individual ‘Arbeitsbezogenes Verhaltens und Erlebensmuster’, Kurzform 44 (Pattern of Work-related Coping Behavior, short-form 44; AVEM-44 [45]) and its change over time, audio samples with progressive muscle relaxation exercises [46], the GSA worksheets and a moderated patient forum. All these features are accessible for 12 weeks. The internet-based therapeutic interventions are performed by two trained psychologists with regular supervision. With the participants’ consent, all blog contacts between therapist and patient are fully documented in the database, allowing subsequent comparisons and analyses concerning the contents.

The control group receives regular e-mail reminders to use selected and online-deposited information about stress management and coping (physical activity, relaxation, healthy diet and sleep hygiene) over the same period of time. All information is permanently downloadable for participants of the control group for the duration of 12 weeks. Overall there are six e-mail reminders, that is, one every fortnight, beginning with a universal reminder, followed by reminders for the five specific topics on stress management and coping.

As shown in Figure 1, the time points of assessment by self-rating scales are at the beginning (T0) and the end (T1) of inpatient rehabilitation, at the end of the aftercare (T2 = 3 months after T1) and 9 months later (T3 = 12 months after T1). Additional somatic and psychological measures, documented during inpatient rehabilitation, are assessed at T1.

Depending on the technical equipment in the rehabilitation centers, T1 assessment is conducted online or paper-pencil. All following data are collected internet based, unless the patients do not participate within two weeks. In this case, the patients get paper-pencil questionnaires by mail. The online assessment takes about 45 minutes, comparable to the paper-pencil assessment. Patients get computer-generated feedback following each completed online questionnaire, describing the participants’ individual work-related behavior and experience pattern and its change over time, measured by the AVEM-44 [45]. The feedback is intended to serve as an incentive for participating in the survey.

In this study we evaluate the acceptance and efficacy of an internet-based aftercare program for three different indications of medical rehabilitation (at the participant level), with the following hypotheses:

The primary outcome measure is a risk factor for early pension, measured by the screening questionnaire SIBAR, a score ≤8 indicating a decreased risk for premature pension. The primary endpoint will be the assessment at the end of the internet-based treatment (T2, that is, 3 months after inpatient rehabilitation). Secondary outcome measures include the subjective prognosis for work and capability, also measured by the SIBAR, and scales for the work-related behavior and experience pattern (AVEM-44 [45]), physical and mental health (Patient Health Questionnaire [47], Generalized Anxiety Disorder 7-item Scale [48], Short-Form 12-item Health Survey [49]), social support (Berlin Social Support Scale [50]), therapeutic alliance (Helping Alliance Questionnaire [51], Helping Alliance Questionnaire II [52]) and self-developed items measuring patients‘ satisfaction with the aftercare program. Long-term effects of the primary outcome measure will be analyzed 9 months after termination of the internet-based aftercare at T3. Except for the Helping Alliance Questionnaires, only applied in the intervention group, all instruments are used in both randomized arms.

At present, there are no reports of reliable effect sizes for internet-based interventions aiming at the improvement of vocational strains through online aftercare interventions. We therefore made conservative calculations with an expected small to moderate effect size (d = 0.30), a statistical power of 0.80 and alpha of 0.05. A total sample size of n = 190 will be necessary to gain significant results (calculation with GPower 3.0.10 for analysis of variance with fixed effects, special, main effects and interactions).

Patients are randomized in clusters - the cluster size is set at two to twelve participants. For each clinic, 30 blocks have been generated, each consisting of both groups in random order. Altogether, there is a maximum of 60 clusters per clinic with an equal share of intervention and control groups. To adjust sample size calculation for clustering, we multiplied the needed sample size by the design effect (1+(m-1)ρ, with m = average cluster size and ρ = intracluster correlation coefficient) as suggested in the CONSORT statement extension for cluster randomized trials [53]. With an anticipated average cluster size of six participants and an intracluster correlation coefficient of ρ = 0.30, as reported by Campbell et al. for comparable studies [54], we calculated a design effect of 2.5. By multiplying our needed sample size of n = 190 with the design effect we get a minimum sample size of n = 475 needed for robust statistical analyses. Even in case of a dropout rate of 40% as reported in a recent aftercare study [21], the anticipated number of n = 800 (n = 100 per clinic and condition) will provide statistically confirmed data.

With the help of the computer software Research Randomizer [55], a block randomization has been conducted by the Study Centre of Mental Disorders. Stratified according to the four clinics to achieve an equal share of intervention and control groups, the lists have been assigned to the trial sites by a researcher not involved in the study.

For the primary analyses, analysis of covariance will be used to compare SIBAR risk scores 12 weeks after the discharge of inpatient rehabilitation (T2) between the internet-based aftercare program (intervention group) and the control group, with covariates for rehabilitation clinic and baseline SIBAR risk score (T1). Cohen’s effect sizes will be calculated. Intention to treat analyses as well as completer analyses will be conducted. Replacement strategies of missing values will be discussed after assessing the pattern of the missing value structure.

For the secondary analyses, self-report questionnaires (AVEM-44; Patient Health Questionnaire; Generalized Anxiety Disorder 7-item Scale; Short-Form 12-item Health Survey; Berlin Social Support Scale; Helping Alliance Questionnaire) will be analyzed by mixed models with repeated measurements. It is expected that the SIBAR risk score and the other self-report measures will not be linear across time. Therefore, these measures will be evaluated by a linear model with fixed effects for treatment and rehabilitation center and time. There will be an indicator variable for the post treatment measurements and the SIBAR risk score at baseline will serve as a covariate. All analyses will be conducted on a two-sided level of significance of α = 0.05.

The study protocol and the final version of the written informed consent form were approved by the Ethics Committee of the Federal State of Rhineland Palatinate (Germany), which is responsible for the Principal Investigator (Ref. No. 837.415.10[7424]) and by the ethics committees responsible for the cooperating rehabilitation clinics.

The procedures set out in this protocol pertaining to the conduct, evaluation and documentation of this trial were designed to ensure that all persons involved in the trial abide by Good Clinical Practice and the ethical principles described in the current revision of the Declaration of Helsinki. The trial will be carried out in keeping with local legal and regulatory requirements.

Before being admitted to the clinical trial, patients must consent to participate after the nature, scope and possible consequences of the clinical trial have been explained in a form understandable to them. The patients must give written informed consent to participate in the study, including their consent to publish.

We expect a very low risk for adverse events. Nevertheless, participants have the possibility to contact the staff at the study center via e-mail or on a mobile help number with the guarantee of a response during working hours (8.00 a.m. to 5.00 p.m.) on weekdays. On weekends and at all other times, patients are referred to a Germany-wide crisis telephone number, which is indicated on our internet platform. All adverse events reported by the participants or detected by the online therapist or respective study staff will be collected during the trial and must be documented in the case report form. The clinical course of the adverse event will be followed by the principal investigators (MEB, RZ), who will contact the participants in case of a crisis for further diagnostics via telephone or for necessary referrals to practitioners near the participant.