Possible associations between callers’ degree-of-worry and their socioeconomic status when contacting out-of-hours services: a prospective cohort study

The aim of this paper is to examine if low SES, marital status and non-Western ethnicity are associated to high DOW of callers contacting OOH services.

A prospective cohort study. Data collected was linked with data on individual SES from Statistics Denmark.

The Copenhagen Emergency Medical Services, the Capital Region of Denmark and the OOH services are integrated into one organisation and can be reached through two telephone numbers; 112, the Copenhagen Emergency Medical Services for life-threatening emergencies and 1813, the OOH services for acute, non-emergent medical calls. The OOH services are available from 4 pm to 8 am on weekdays and around the clock on weekends and holidays. Individuals may also call the OOH services for a referral to an emergency department, if they cannot get in touch with their GP during regular working hours. All access to acute care is pre-assessed by telephone triage. Annually, approximately one million calls are handled by call handlers (nurses/physicians) who triage the caller to self-care, GP, home visit, face-to-face assessment/consultation at a hospital, or direct hospitalisation [26, 27]. Call handlers use a criterion-based electronic triage tool, which is locally developed and not validated.

Callers to the OOH services from January 24th to February 9th, 2017, were invited to participate in the study. Inclusion criteria were age ≥ 18 years of the patient and being able to link patient’s unique PIN (personal identification) number to Statistics Denmark. Exclusion criteria were lack of consent, not being the patient themselves or a close relative/friend to the patient, repeat callers, not having a permanent PIN and not rating DOW (see Fig. 1). Via telephone survey, participants electronically reported their DOW on a five-point scale (1 = minimal worry to 5 = maximum worry) before being put through to a call handler. Callers’ DOW was stored in their electronic medical record. An in-depth description of this data collection has been published previously [13].

Data was derived from two different sources:

The socioeconomic variables were categorized as follows:

Age was categorized by 18–30, 31–40, 41–50, 51–60, 61–70, 71–80 and > 80 years.

For statistical purposes, DOW was dichotomized into two groups: low (DOW = 1–3) and high (DOW = 4–5), based on the association between DOW and hospitalization, where DOW = 4–5 showed significantly higher odds for hospitalization than DOW = 1–3 [13].

Associations between education, annual household income, marital status and ethnicity and high DOW were analysed separately, using logistic regression models adjusted for gender and continuous age. Results were reported as odds ratios (OR) and 95% confidence intervals (95% CI).

A total of 38,787 callers were invited to participate in the study. Of these, 25,885 callers declined participation, 699 were not patients themselves or a close relative/friend of the patient, 771 were repeated callers, 73 had a temporary PIN [29] and 19 did not rate their DOW. This resulted in inclusion of 11,340 callers, where only adult callers (≥18 years) (n = 4377 excluded) that could be linked by their unique PIN to registers at Statistics Denmark (n = 94 excluded), were included in the present study, resulting in analysis of a total of 6869 callers (see Fig. 1).

DOW was rated as high by 38.2% of the 6869 callers. The study population consisted mainly of females (58.3%), with the proportions of females being similar in both the low (57.9%) and high (59.0%) DOW category. The mean age of those reporting a high DOW was 51.6 (standard deviation (SD) 20.9) years compared to 43.4 (SD 18.7) years for low DOW. The proportion of callers reporting high DOW increased with age; with 29.5% of callers aged 18–30 years reporting high DOW compared to 57.5% of callers aged > 80 years.

Educational level was registered for 6573 callers, with most callers (2820; 41.1%) having a middle educational level. Information on household income was available for 6129 callers, with the mean annual household income (DKK 266,986 (SD 209,711)) falling in the middle category. The majority of the callers (3847; 56%) was married or cohabiting. Finally, the majority (83.1%) of callers were of Danish ethnicity, whereas only 4.2% were of non-Danish Western ethnicity and 12.7% were of non-Western ethnicity (Table 1).

Low educational level was associated with high DOW (OR 1.5, 95% CI 1.3 to 1.7), compared to middle educational level. Whereas, a high educational level was associated to reporting a low DOW compared to middle educational level. Low annual household income was associated with high DOW (OR 1.4, 95% CI 1.2 to 1.6), compared to middle annual household income. Whereas high annual household income was associated to a low DOW compared to middle annual household income. Being single was associated with high DOW (OR 1.3, 95% CI 1.1 to 1.5) compared to being married or cohabiting. Non-Western ethnicity was associated with high DOW (OR 2.9, 95% CI 2.5 to 3.4), compared to Danish ethnicity, but there was no significant association with non-Danish Western ethnicity (see Table 2).

Previous studies have demonstrated an association between socioeconomic variables (individuals’ SES, marital status and ethnicity) with morbidity and self-rated health. Low SES (low education [16, 30] and low annual household income [31, 32]) are associated with high morbidity and low self-rated health [33]. The same is true for marital status as single [34] and in Denmark, individuals of non-Western origin also have higher rates of morbidity [25]. There is no direct explanation as to why low SES, marital status as single and non-Western ethnicity are associated with high DOW. However, it could be speculated, whether the association between these socioeconomic variables and high DOW is caused by the general social gradient in health care both in detection as well as treatment of illness [35‐37].

DOW was rated as high by 38.2% of the participants. This minority may be explained, as callers with life-threatening emergencies are prompted to contact the Copenhagen Emergency Medical Services and only the OOH services for less urgent situations. The majority of the participants (61.8%) reported a low DOW. This coincides with reports showing that approximately 50% of callers to OOH services are referred to selfcare or their GP the next day [38]. One could fear a possible risk of call handlers overlooking serious illness if patients report a low DOW. However, DOW is meant as an additional tool in the triage process by encouraging patients’ participation, increasing patient-centred communication and thereby aiding call handlers in determining urgency. It should not stand alone. Furthermore, it is yet to be examined whether patients’ DOW when calling OOH services correlates to the DOW of a close relative/friend when calling on behalf of a patient. However, the same understanding applies with the use of DOW from a close relative/friend as from the patient herself, by encouraging caller participation, increasing caller-centred communication and giving callers the opportunity to voice their concerns.

The main strength of this study is the use of data regarding SES, marital status and ethnicity obtained from registers, thereby minimizing information bias of self-report of socioeconomic variables. Furthermore, participants’ DOW was obtained when they were seeking help before being put through to a call handler, thereby limiting the influence of the triage result. Limitations include a possible risk of selection bias among callers who declined to participate, however, selection bias was assessed in a previous study [14] and no differences between participants and non-participants regarding gender, age, reason for contacting OOH services and triage response were found. Non-participation may, however, be more pronounced among callers with low SES due to low health literacy, callers with non-Western ethnicity due to possible language difficulties [39] and callers who were extremely distressed when contacting the OOH services. One may also argue that, as callers with low SES, marital status as single and non-Western ethnicity have higher morbidity and lower health literacy rates [19, 40], naturally they would more often have an increased sense of medical urgency, an association that previously been shown [41], leading to a high DOW. This possible bias, however, remains to be explored. The data includes missing variables of education and household income. The data on highest attained educational level is a little below the general coverage of the register (95.7% vs 96.4%) and within the category of missing information an association with high DOW is observed.

Previous studies have shown that the use of OOH services is significantly higher in areas with low SES [42] and it is therefore important to understand callers’ motivation for help-seeking in this context. As call handlers in OOH services cannot rely on visual cues in order to assess urgency during telephone triage, they must solely rely on verbal information given by callers, which may obscure their assessment of urgency. Moreover, as callers are unknown to the call handlers, their full medical history, SES and level of health literacy is not necessarily known. Experts within the field of emergency medicine decision-making recognise, that call handlers’ perceptions may be influenced by mental shortcuts and an unconscious integration of preconceptions and stereotyping [2, 43, 44]. The clinical relevance of this study does not lie in the statistical significance, but in the approach to the patient, and serves as a reminder that the symptoms the clinician deduce from a patient is the interpreted narrative. Incorporating DOW as an additional permanent triage tool may aid the triage process by encouraging patient participation, increasing patient-centred communication and giving callers the opportunity to voice their concerns, which may especially be advantageous to callers with lower health literacy or challenged verbal ability. Callers’ DOW could thereby, act as a mental forcing strategy, potentially debiasing the call handler’s perception of the caller. According to the Common-Sense Model of Self-Regulation (CSM) by Leventhal [45], a widely recognized theoretical framework, patients’ perceptions of their illness, is based on prior experience, personal beliefs, discussions with others and cultural understandings [46] and a relation between a patient’s illness representation and self-evaluation of urgency or DOW has previously been shown [12]. If the callers’ DOW, therefore, does not correspond to the call handler’s sense of urgency, the call handler may ask more in-depth questions, leading to a better understanding of the patient’s illness perception, especially in relation to DOW, possibly influencing and correcting the call handlers sense of urgency. Acknowledging the patient’s DOW as valuable triage information will enable elicitation, attention to, and understanding of the individual meaning and significance of patient’s symptoms [47].

Thus, when call handlers are presented with a high DOW from a caller, especially if this does not correspond to the call handler’s perception of urgency, an awareness of the significant association to various socioeconomic variables and high DOW, should urge call handlers to a more thorough questioning of the caller for possible severe illness, as this population group may have a higher risk of serious illness. This increase in patient-centred communication may aid the call handler in determining the correct type of health care needed, thus ensuring patient safety. Furthermore, the association of callers’ SES, marital status and ethnicity with DOW when contacting OOH services adds to the growing awareness of the socioeconomic inequality in health and healthcare and the continuing necessity of minimizing this gap. This is a new area of research and this study gives direction for future research to further strengthen the evidence.

There is a social and ethnic gradient in self-reported DOW when contacting an OOH, with low SES, marital status as single and non-Western ethnicity being associated with high DOW.