Aim
The aim of this study was to explore patient motives for seeking acute healthcare at OOH-PC and the EMS and to investigate motives associated with contacting each of these two healthcare service providers.
Design and setting
We conducted a cross-sectional observational study to explore patients’ motives for contacting OOH care by sending a questionnaire to patients who had contacted OOH-PC and the EMS. Data was collected in two Danish regions, the Capital Region of Denmark in Copenhagen and the Central Denmark Region, during a two-week period in February-March 2015.
All citizens with fixed abode in Denmark are listed with a GP and have access to the public (tax-funded) healthcare system free of charge. GPs serve as gatekeepers to secondary care and are usually available on weekdays from 8 am to 4 pm. Denmark is divided into five regions; each of these regions is responsible for organising healthcare in their own region. Healthcare is provided by primary care (both daytime and OOH-PC), the EMS and secondary care (e.g. hospitals, EDs). Referral from either primary care or the EMS is required before an ED visit or hospital admission. Therefore, we did not include the ED.
In the Central Denmark Region, OOH-PC is organised by GPs in large-scale cooperatives (GPCs). GPs perform telephone triage and deal with the presented problem by giving telephone advice or by referring the patient to a subsequent face-to-face consultation [24]. In the Capital Region of Denmark in Copenhagen OOH-PC is an integrated part of the EMS, with medical helpline 1813 (MH-1813) serving as a dedicated entrance for non-urgent cases. Nurses perform the triage; they are supported by a computerised decision-support tool and the opportunity to consult a doctor (or hand over the call), but these doctors may also answer direct calls. Patients receive telephone advice or are referred to a face-to-face consultation.
In both regions, the EMS consists of an emergency medical coordination centre (EMCC) handling all 1-1-2 emergency calls. The EMCC is staffed by different types of healthcare professionals (nurses, paramedics and doctors for supervision), who asses the urgency level and decide on the suitable response, as indicated by the criteria-based dispatch protocol named the Danish Index for Emergency Care [25]. This protocol states 37 main dispatch criteria (symptoms) and divides calls into 5 levels of emergency.
Study population
We included patients who contacted OOH-PC or the EMS outside office hours (i.e. weekdays from 4 pm to 8 am, entire weekends and bank holidays). The first contact with a healthcare professional within the study period was included for each patient. If a patient had a follow-up contact with the other OOH healthcare service provider within the study period, we included only the first contact. Exclusion criteria were: contact during daytime, death at the time of dispatching questionnaires, address protection, living in an institution, tourists and other citizens with an invalid personal identification number (PIN) [26] and participation in one of the pilot studies. Patients aged 13-18 years were also excluded for confidentiality reasons. Moreover, for EMS contacts, we excluded patient transport planned in advance and requests for an acute ambulance by healthcare professionals. A recorded message on the telephone waiting line informed patients calling the GPC and MH-1813 about the ongoing research project, and callers were given the opportunity to decline participation by pressing ‘9’.
Development of questionnaire
First, a literature search was conducted, and existing questionnaires on patient motives were studied, resulting in an overview of factors related to decision-making in patients and prevailing motives for contacting OOH care. Next, these factors and motives were categorised into a model for decision-making when contacting OOH healthcare services. The model was based on Andersen’s Behavioural Model [27] and adapted to the Danish healthcare system. Several internal research meetings and an external expert feedback round were held, resulting in the final questionnaire (Appendix 1). Motives were measured by 26 predefined statements relating to the decision to contact OOH care (Appendix 2). Respondents were asked to rate, on a 5-point Likert scale, the importance of each statement for the decision. Motives were grouped into: ‘own assessment and expectations’, ‘barriers and benefits’, ‘previous experience and knowledge’, and ‘needs and wishes’. We tested the questionnaire twice in the GPC waiting room and interviewed four patients in a general practice waiting room to ensure clarity and validity. Moreover, three small-scale pilot studies were conducted to enhance clarity, increase the response rate and enable a power calculation.
The final questionnaire included questions on patient characteristics, the health-related problem and the patient's motives for contacting OOH care. Patient characteristics included: age and gender, decision maker (patient himself/herself, family member, other known person or unknown person), ethnicity and marital status. Questions about the health-related problem included: main problem and duration. In addition, we included information extracted from the patient registration systems of the OOH healthcare service providers: date and time of contact, patient’s PIN, type of contact and urgency level (only for EMS contacts). The PIN was used to calculate age and gender, search for duplicates and check the patient’s status (possible death) before sending the questionnaires.
Data collection
A power calculation showed that we needed a study population of 400 respondents per healthcare service provider, each consisting of two units, to be able to detect a 10% difference in the importance of motives between the OOH-PC and EMS, as well as between the GPC and MH-1813. Having obtained a response rate of 40% in our final pilot study, we aimed to send out 1,000 questionnaires per healthcare unit. As we also aimed to compare motives for contacts regarding children and adults, we selected 1,000 patients <13 years and 1,000 patients >18 years for both the GPC and MH-1813. We selected 1,100 patients rather than 1,000 per EMS unit, as we expected a lower response rate and more exclusions due to high numbers of bystander calls and incorrect PINs. The data collection lasted one week for the two OOH-PC service providers (GPC and MH-1813) and two weeks for the EMS due to differences in number of weekly patient contacts.
Data on calls were received twice a week, and questionnaires were dispatched within four days after the relevant OOH contact to ensure vivid recall of contact details and the decision-making process. We randomly selected contacts from each healthcare unit, and the patient’s address and status (alive/deceased) was verified in the Civil Registration System. Invitation letter and paper questionnaire were sent to patients aged >18 years and registered guardians for patients aged <13 years, including a link and login credentials to a web-based version of the questionnaire. One reminder was sent after two weeks.
Statistical analyses
Descriptive analyses were performed to identify the main characteristics of contacts and respondents, stratified by healthcare service. Motives were dichotomised into ‘not important’ (‘not relevant’, ‘no importance’, ‘little importance’, ‘some importance’) and ‘important’ (‘important’ and ‘very important’), and the percentage of importance per motive was estimated for each of the two healthcare services. Aiming to identify motives for contacting the EMS (as opposed to contacting OOH-PC) and to obtain risk ratios (RR), we applied the modified Poisson regression model for all contacts and stratified for children and adults [28]. The resulting RRs were presented along with 95% confidence intervals. Moreover, a non-response analysis was conducted. Stata statistical software, version 14, was used (StataCorp LP, College Station, TX, USA).