Summary of findings
Our empirical studies add significantly to current knowledge about frequent callers to telephone helplines. Most of them are larger in scale and more rigorous than their predecessors. They explicitly consider differences between frequent callers and one-off or episodic callers. Collectively, they allow us to triangulate our findings and consider whether there are some findings that appear to be consistent across studies with different methodologies and data sources. They provide a similar picture of frequent callers, and thus give us confidence that the profile that we have identified is valid.
Several key findings stand out. Frequent callers are relatively few in number but they account for a substantial proportion of calls. They have a heavy reliance on helplines, perhaps because they are isolated and have relatively few social supports. They are by no means just “time wasters”, however; they have high levels of need, as evidenced by the fact that they have major mental health problems (including anxiety, depression and suicidality) and are often in crisis. They also make use of other services for their mental health problems, including GPs, allied health professionals (e.g., psychologists), psychiatrists and emergency departments. The circumstances under which frequent callers make use of telephone helplines vary, but current service models are not meeting their needs and are reinforcing their calling behaviour.
Strengths and limitations
The key strength of this body of work is that it brought together four separate empirical studies to address identified gaps in the existing literature. These studies utilised a range of data sources and employed a variety of different methods, allowing us to triangulate our findings and draw conclusions with a degree of certainty that was not previously possible.
Individually, of course, each of our studies had limitations and these have been acknowledged elsewhere [
19‐
22]. Mostly, these related to the fact that, with the exception of our survey/interview study, we were reliant on routinely collected data (as with our analyses of Lifeline calls data) or data collected in the context of other studies whose original emphasis was not on frequent callers (as with our analyses of data from the
diamond study and the NSMHWB). This sometimes meant that particular variables were collected in a way that was not always ideal for our purposes and/or that the numbers of frequent callers were small. This in turn meant that we were not always able to operationalise frequent callers as those who called 20 times or more within a month and/or had to aggregate callers into imperfect groups. With the
diamond study, for example, we had to deem those who called once a week or more as frequent callers because of the way the data were collected. With the NSMHWB, we were restricted to looking at those who called more than once in a 12 month period (termed repeat callers) as a group, because the numbers of individuals who called at higher rate than this were too small for meaningful analysis.
The findings from our empirical studies show that the current model of service delivery is not working for frequent callers. Features of the model—e.g., the fact that callers remain anonymous and can make unlimited calls—reinforce their calling patterns. An alternative model is required that better serves the needs of both frequent callers and other callers who use telephone helplines episodically or in a one-off manner. We propose a new model here.
The proposed model is guided by the following principles:
-
The model addresses a problem that is about behaviours (frequent calling) not individuals (frequent callers). The existence of these behaviours provides evidence that these callers’ needs are not currently being met. These behaviours present a challenge for telephone helplines but it may be possible to “turn them around”. Those engaging in these behaviours may benefit from a different approach.
-
The model is non-judgemental. It recognises that frequent callers have high levels of genuine need. They have complex mental and physical health problems and a range of social issues, and experience crises that may not be quickly resolved but instead may be more ongoing in nature and may be heightened by specific triggers or at times of stress and anxiety. They are isolated, but that this is not the sole reason for their calling patterns.
-
The model recognises that individuals are different and therefore offers flexibility and choice. Some frequent callers will “opt in” and take advantage of the new model, whereas others may prefer not to.
-
The model articulates clear roles and responsibilities for frequent callers who do choose to use the new model of service delivery. It empowers frequent callers by involving them in early decisions about their ongoing care (e.g., goals) and commits them to calling at agreed times.
-
The model acknowledges that frequent callers are accessing a range of other services in addition to telephone helplines. It capitalises on this, and promotes collaboration between telephone helplines and these other services wherever possible and appropriate.
The features of the model, and some of the conditions that would necessarily have to underpin them, are described below.
Dedicated and specially trained TCSs
Under the proposed model, a sub-group of TCSs would be dedicated to taking calls from frequent callers. They would receive additional training for this, over and above the normal training offered to TCSs. They would also have ongoing opportunities for continued professional development, one-to-one supervision sessions, debriefing, and peer support.
The training and other support offered to these TCSs would focus on equipping them to deal with mental health issues (particularly anxiety, depression and suicidality) and social issues (particularly loneliness and isolation), and on the overlap between these and physical health problems. It would also prepare these TCSs to deal with some of the complexities underpinning frequent callers’ behaviour, including, for example, attachment issues.
The result would be that the telephone helpline would have a cadre of highly skilled TCSs with specific expertise in dealing with frequent callers. Their role would be akin to that of a mental health counsellor in other community settings. These TCSs would cover a significant number of shifts in any given week, and be available at times that frequent callers are particularly likely to call (e.g., at night).
Consideration might be given to whether these TCSs should be retained on a paid basis, rather than as volunteers, and whether individuals with some tertiary or equivalent training in dealing with mental health issues might be preferred over lay people.
An integrated, tailored service
Frequent callers would be allocated to one of these specialised TCSs who would develop a rapport with them, establish rules about the timing and duration of their calls, and help them work towards clearly defined goals. The caller and the TCS would reach an agreement about how often the caller could use the service, the type of care he or she should expect to receive, and what to do in the case of an emergency. The TCSs would provide a more intensive, high level of counselling than the standard telephone helpline service.
The TCSs would work with these callers to agree on well-articulated management plans, based on their history and presenting issues. Then, in accordance with these plans, they would assist them to develop strategies for dealing with their various mental health and social issues. They would also explore callers’ relationships with the telephone helpline and guide them towards more adaptive relationships by modelling secure attachment behaviour [
37], setting defined boundaries, and articulating clear consequences of breaching these boundaries. They could potentially draw on a range of therapeutic approaches. For example, cognitive behavioural therapy (CBT) [
38] could be used to help individuals identify and address maladaptive cognitions and behaviours that motivate their calling behaviour. Alternatively, acceptance and commitment therapy (ACT) [
39] could be used to assist callers to come to terms with unpleasant thoughts that drive them to call telephone helplines, and develop strategies to reduce their likelihood of acting on these thoughts.
Calls would take the form of a series of sessions at which particular issues might be discussed or particular strategies might be taught. At each call, the TCS would ask the caller how things are going with issues that were discussed on the previous call, and would follow up with any homework tasks that might have been set. The TCS would acknowledge previous conversations but would guide the caller away from ruminative thought processes.
The model would also draw on newer technologies that might facilitate greater levels of care without being too resource intensive (e.g., the issue of loneliness might be addressed through a facilitated webinar group that meets to discuss topics related to wellbeing, and symptoms of anxiety and depression might be addressed via mobile applications or interactive websites, of which there are many examples [
40]). Of course, approaches involving newer technologies might not suit all callers.
Linkages to other services
The model might be thought of as part of a stepped care process. Some callers might only use this service, whereas others might be assisted to “step up” to other services, beginning with primary care and, if necessary, moving up to specialised mental health services. This would require good linkages between the telephone helpline and other service pathway elements.
The model recognises that frequent callers are likely to already be using a range of these other services, including GPs and mental health specialists. It is not about creating new linkages but improving the quality of existing ones, reducing reliance on multiple providers (e.g., several GPs), and fostering consistent approaches. For example, there would be instances in which the TCS might work with the caller and his/her GP on a shared care plan.
A seamless triage system
For the model to work, a seamless triage system would need to be put in place. Frequent callers would be identified by a variety of means (e.g., through the helpline’s telephony system “flagging” their telephone number, or through cues that they give when they introduce themselves). Once identified, they would be offered the opportunity to speak to one of the dedicated and specially trained TCSs. If they took this up, they would be put through to the TCS who would explain the service to them in more detail and invite them to make use of it. For those who chose to “enrol” in the service, this would act as the first session.
Rules of engagement
The ongoing relationship between frequent callers and specific TCSs would require that callers relinquish their anonymity and give their names and contact details. The TCSs would also be required to give their names, although they might use pseudonyms. This open use of names would foster rapport and trust, and would make for more “normal” conversations. It would also be necessary for practical reasons, in order to ensure that the caller could always make contact with his or her allocated TCS.
As noted above, frequent callers making use of the integrated, tailored service would enter into an agreement with the TCS about the timing of their calls. Effectively, they would be calling at agreed appointment times, some of which would be made available after business hours. Callers in an acute crisis could call outside these times, but would be diverted back to the regular service. This arrangement would need to be explained to callers in a manner that did not encourage further frequent use of the helpline, but empowered them to move towards recovery. Further work is required to determine the optimal approach to restricting frequent callers’ calls in this way, and it is likely that the solution will be different for different callers. One option, for example, might be to taper the calls over time, initially allowing callers to call daily, then 2–3 times per week, then weekly, then fortnightly, then monthly etc.
Although the caller would be paired with an individual TCS, the fact that there would be a critical mass of these specialised workers would mean that callers could potentially shift from one TCS to another if they felt that would be beneficial. Equally, TCSs might cross-refer callers, because it is likely that individual TCSs might develop expertise for dealing with particular types of frequent callers, and effectively become super-specialists.
Getting the balance right
The proposed model would need to be implemented in a way that ensures that it does not amplify frequent callers’ reliance on telephone helplines. It should: be viewed as an intensive but time-limited service that helps frequent callers to move on with their lives; provide an opportunity for callers to develop a meaningful connection with a specific TCS without encouraging further dependency; and be seen as an alternative to their regular use of the telephone helpline, rather than as an adjunct to it (although, as noted above, it will be necessary to allow callers to make a standard call to the helpline in an emergency). The nomenclature around the model would also require careful thought; it would need to be non-stigmatising without “rewarding” frequent callers by offering them a specialised service.
Suggested next steps
We would suggest that the next step in the process of dealing with frequent callers is to further refine the proposed model, testing the concept out with key stakeholders. There may be elements that are missing, or existing elements that are seen as unworkable. These stakeholders should include frequent callers themselves, TCSs, supervisors and managers, and representatives from primary care and specialist mental health services. Frequent callers would clearly have a view as to whether this sort of model would address their needs, and our survey/interview study suggests that they are very willing to participate in relevant information-gathering exercises. TCSs would provide valuable insights into the extent to which the model might alleviate some of the stresses associated with dealing with frequent callers, as well as views about whether the model might introduce new issues (e.g., by creating a parallel system of service delivery). Supervisors and managers would provide input from organisational/systems perspectives, and would be able to comment on the workability of the proposed model. Primary care providers (particularly GPs) and specialist mental health providers would have views on how best to formalise collaborative relationships.
One approach to involving stakeholders in the model’s refinement might be to use a co-design methodology [
41]. This has been used in other areas of health and social care to reconfigure service systems to better address needs. It emphasises the experiences of stakeholders—particularly users—with the current system. It recognises that service users are not passive recipients of services but instead are integral to ideal models of care.
Once input from frequent callers and other stakeholders has been received and the model has been further refined, it should be tested in a controlled way. This is crucial because the model is not without risks; although it is intended to reduce the reliance of frequent callers on telephone helplines and free up resources for other callers, it is possible that it could have negative impacts. We would recommend piloting the model in a few telephone helpline centres in the first instance, and evaluating it in a methodologically rigorous way. Ideally, this would involve a randomised controlled trial in which frequent callers were randomly allocated to receive the tailored, integrated service or to receive usual care. The trial would consider both the effectiveness of the new service (i.e., its achievement of benefits for frequent callers) and its cost-effectiveness (i.e., weighing these benefits up against the cost of implementing the new service). Careful examination of unintended consequences would also be important. For example, it would be necessary to monitor the total number of calls made by frequent callers allocated to the new service to ensure that they were not in fact making greater use of the given telephone helpline (i.e., using the new service and the regular service in tandem). In addition to focusing on outcomes, the trial should also monitor the processes associated with the new service (e.g., the way in which each of its elements is operationalised) and impacts (e.g., the effects the service has on the telephone helpline as a whole and its TCSs in particular). It should also have an emphasis on the amount and level of training required by the dedicated TCSs.
Lifeline has expressed interest in exploring the potential of the model further, following the sorts of steps we have described above. If the model proves to be successful in this context, consideration might then be given to how to optimise it and ensure that it is replicable and scalable in other services, either in Australia or overseas. Implementation science literature suggests that consideration of the different operational frameworks and different skill sets and experiences of TCSs will be important here [
42].