- Research article
- Open Access
- Open Peer Review
Reasons for encounter by different levels of urgency in out-of-hours emergency primary health care in Norway: a cross sectional study
BMC Emergency Medicinevolume 17, Article number: 19 (2017)
Frequencies of reasons for encounter (RFEs) in emergency primary care out-of-hours (OOH) services are relevant for planning of capacities as well as to target the training of staff at casualty clinics. We aimed to present frequencies of RFEs in the different organ systems, and to identify the most frequent RFEs at different urgency levels.
We analyzed data on RFEs in Norwegian OOH services. International Classification of Primary Care (ICPC-2) RFE codes were recorded in all contacts to eight representative OOH casualty clinics in 2014 and 2015 covering 20 municipalities with a total population of 260 196. Frequencies of each ICPC-2 chapters and groups of ICPC-2 codes were calculated at different urgency levels.
Musculoskeletal, respiratory, skin, digestive and general and unspecified issues were the most frequent RFE groups. Fever was the most frequent single ICPC-2 RFE code, but was less common among the most urgent cases. Abdominal pain was the most common RFE in patients with yellow urgency level (urgent), and chest pain dominated the potentially red (potentially life threatening) cases. There was less variation in the use of ICPC-2 with increasing urgency level.
This study identifies important differences in RFEs between urgency levels in the Norwegian OOH services. The findings provide new insight into the function of the primary health care emergency services in the Norwegian health care system, and should have implications for staffing, training and equipment in the OOH services.
The organization of out-of-hours (OOH) emergency health care services varies considerably between countries, and this may lead to differences in what medical conditions the OOH services handle. In most countries, the population have direct access to emergency rooms in hospitals, although primary care services are available. In the public health care system in Norway, strict gatekeeping is fundamental, even for emergency health care . In principle, even the most life-threatening conditions have to be assessed by primary health care doctors before admission to a hospital. Consequently, all local municipalities have to ensure access to OOH emergency primary health care (Norwegian “legevakt”) at all times for their inhabitants and all other individuals present in that area . Although the OOH services should be designed to handle the most urgent cases, the services generally have a low threshold for taking in patients. Thus, most of the activity in OOH casualty clinics involves patients with non-urgent conditions. Data from a sentinel network for monitoring the activity of OOH services in Norway indicate that almost three in four contacts were classified as “green” or not urgent and only 3 percent were “red” or potentially life threatening . In order to sufficiently staff and equip an OOH clinic according to demand, information about the amount and type of patients is crucial. Also for the development of standardized qualification requirements for casualty clinic staff, it is also important to know which conditions OOH nurses and doctors should expect at different urgency levels, and how often.
Demographics (age, gender) of OOH patients and the utilization of OOH-services at different times in Norway is relatively well characterized . Statistics on diagnoses on reimbursement claims in the Norwegian OOH services is available , but its usefulness is limited by extensive use of inaccurate and general diagnose codes and the fact that the diagnosing is partly based on financial incentives. The reasons for encounter (RFE) given by patients on first contact is probably more relevant than the doctors diagnoses when planning capacities of future OOH clinics. RFEs give a better picture of the amount and types of unresolved cases that can be expected, which is important to the function of an emergency health care organization, particularly with regard to the gatekeeping performed by OOH nurses.
Knowledge about RFEs is important to characterize the role of emergency OOH services in primary health care in urgent and non-urgent situations. RFEs in OOH services have been studied previously [5, 6], but not in light of urgency level. We wanted to identify frequencies of RFEs in the different organ systems, and aimed to identify the most frequent RFEs at different urgency levels in emergency primary care.
We performed a cross sectional study based on data from the “Watchtower project”, a sentinel network of representative OOH emergency primary health care activity in Norway. The development and implementation of this project has been described in detail elsewhere .
Data was missing from one of the Watchtower OOH districts for the first three months of 2014 for technical reasons in connection with a reorganization. We chose to exclude the three municipalities from this district to get representative annual RFE rates, leaving Watchtower data from seven OOH-districts covering altogether 20 municipalities with a total population of 260 196, or 5.1% of Norway’s total (2014). The included OOH districts covered 15 810 km2, or 4.9% of Norway’s total mainland area. Average population density was 16.5 people per square kilometer. The population (2014) of the OOH-districts ranged from 4 924 to 93 121.
All patient encounters in the included Watchtower casualty clinics in 2014 and 2015 were included. Data were collected using an online data collection tool developed by the first author using Zoho Creator . For every first contact trained staff (nurse or other) was instructed to record ICPC-2 (International Classification of Primary Care) code according to the official reason for encounter manual . A RFE is the patient’s main complaint upon first contact. Typically, most RFEs are general symptoms or ailments, but distinct diagnoses may also be recorded if explicitly mentioned by the patient. It is important that the nurse records the patient’s own perception of the condition, even though it is unlikely or even wrong. In contrast, ICPC-2 diagnoses require assessment by a doctor, and are often more specific. In this study, we recorded only ICPC-2 RFEs, not diagnoses. ICPC-2 was developed by World Organization of Family Doctors (WONCA), and is used as a medical classification system universally across the Norwegian primary health care. It contains 15 chapters covering the organ systems, and two additional chapters for psychological and social problems. Personnel of all participating OOH districts received training in determining ICPC-2 RFE codes. Only codes for symptoms and complaints (codes -01 to -29), and diagnoses and diseases (codes -70 to -99), were applied. In addition, the staff recorded age and gender of patient, time and mode of contact, home municipality and type first action taken. Urgency level according to the Norwegian Index for Medical Emergency Assistance (Index)  was recorded.
The urgency levels are green (not urgent), yellow (urgent) and red (potentially life threatening). Red responses imply immediate ambulance dispatch and alarming of doctor on call. In yellow responses, the need for ambulance and doctor alarm is constantly assessed, whereas the green responses in many cases can be referred to a normal “in-hours” consultation .
There is a certain overlap between several of the ICPC-2 codes, often across chapters. For example, chest pain may be coded as general chest pain (A11), heart pain (K01), pressure/tightness of heart (K02) or musculoskeletal thoracic pain (L04). The clinical relevance is therefore limited when focusing on single codes. To get a better picture of the RFE repertoire, some of the ICPC-2 RFE codes were classified into 22 groups based on consensus between the authors (Table 1).
The main outcomes are the total RFE frequencies of the different ICPC-2 chapters, both as proportion of total contacts, and as annual rates per 100 000 person-years. ICPC-2 chapter rates for each out-of-hours district were also calculated, and the range was recorded as a measure of variance. Other outcomes include frequencies of ICPC-2 chapters and RFE groups, and the identification of the 20 most frequent ICPC-2 RFE codes in the three urgency levels.
SPSS 22 and Excel 2013 were used to handle data. Counts and relative frequencies of RFEs were calculated in in addition to incidence rates for the total study population. For ICPC-2 chapters, 95% confidence intervals for proportions of relative frequencies were calculated, and the maximum and minimum incidences among the included seven OOH districts were presented. For the 20 most frequent RFE groups and codes, 95% confidence intervals for incidence (Poisson rate) with normal approximation was calculated.
RFE (ICPC-2 code) was not a mandatory field in the data collection tool. We therefore anticipated a relatively high number of cases with missing or unknown RFE. To assess whether there was any selection bias, average age, distribution of gender, urgency level and time of day was calculated as baseline data in cases with and without ICPC-2 RFE codes.
All other fields in the data collection tool were mandatory, but there was an “unknown” option for each piece of information, forcing the operator to actively, and not by default, indicate when information was missing. Although the operators were instructed to record all contacts, some cases unavoidably were not recorded at all. Reasons for this might be that Watchtower recordings had to be down-prioritized in extremely busy periods, or that some of them for different reasons were forgotten. It was not possible to count these omitted cases, but comparisons between extrapolations of the Watchtower data and data based on reimbursement claims, indicate that approximately 15% of cases are missing in the Watchtower database .
In total, 177 053 contacts were recorded from the seven included OOH districts, 86 089 in 2014 and 90 964 in 2015. ICPC-2 RFE code was recorded in 160 215 (90.5%) of cases.
16 738 (9.5%) of the records lacked information about ICPC-2 RFE, and 359 (0.2%) lacked information about urgency level. Of these, 232 (0.1%) had neither ICPC-2 RFE code nor urgency level.
In Table 2 characteristics of patients with recorded ICPC-2RFE code is compared with those with unknown RFE. The differences were small, but the proportion of green cases and the proportion recorded at night was higher among encounters without known ICPC-2 RFE code. There was a higher number of missing ICPC-2 RFE codes when information about gender, urgency level or time of day was also missing.
The distributions of ICPC-2 chapters at different urgency levels are presented in Additional file 1: Table S1, Additional file 2: Table S2, Additional file 3: Table S3 and Additional file 4: Table S4. The number of general and unspecified (chapter A) was high for all urgency levels, especially among the most acute (red) cases, where 34% of RFEs were from chapter A. In red urgency level, 78% of the cases had a RFE from one of the five most frequent ICPC-2 chapters, compared to 65% of green records and 66% of yellow.
Cases with yellow urgency level had the highest proportion of digestive issues (chapter D, 12%). The proportions of neurological (chapter N) and circulatory (chapter K) issues were two- and three-fold higher in yellow cases compared to green respectively, but proportion of chapter H (ear) was less than a fifth.
Eighteen percent of RFEs with red urgency level were circulatory (chapter K) in contrast to only 1.3% of the green and 3.9% of the yellow RFEs. Compared to green cases, patients with red urgency level had a 3 times higher proportion of neurological cases (chapter N). Chapters L (musculoskeletal), U (urological) and F (eye) had a considerably lower proportion of red cases compared with green and yellow. Psychological/psychiatric RFEs (chapter P) were less frequent in green contacts than in yellow and red, where they constituted approximately one in twenty cases.
As can be seen from Additional file 1: Table S1, Additional file 2: Table S2, Additional file 3: Table S3 and Additional file 4: Table S4, there was a considerable variation in incidence rates of the different ICPC-2 chapters from one OOH district to another. The cardiovascular incidence rate ranged from 695 to 1 555 cases per 100 000 person-years. The differences between OOH clinics were most pronounced in the red urgency level, where the variation was 3.7-fold in chapter R (respiratory), 3.8-fold in chapter C (circulatory), 4.7-fold in chapter A (general/unspecified), 5.1-fold in chapter D (digestive), and 5.9-fold in chapter N (neurological).
The 20 most frequent RFEs in the three urgency levels are presented in Tables 3, 4 and 5. These rankings include both RFE groups and ICPC-2 RFE codes that were not assigned a group. The 20 most frequent individual ICPC-2 RFE codes in the three urgency levels are presented in Table 6.
General symptoms, limb issues, respiratory issues and fever were the most frequent non-urgent green RFEs (Table 3). In yellow cases, abdominal pain was the number one RFE (Table 4). Skin injuries constituted 6.6% of yellow encounters compared to 3.6% of green and 0.8 of red. Among the potentially life-threatening cases, chest pain dominated with 25% of RFEs. There were several RFEs that had significant relative high incidences only in the red urgency level (Table 5). Stroke (K90) is one example. When combined with ICPC-2 RFE codes N18 (paralysis/weakness) and N19 (speech disorder), such a stroke-related RFE group would be the fourth most frequent RFE with 249 encounters per 100 000 person-years.
The most frequent single ICPC-2 RFE codes for each urgency level are presented in Table 6. There was less variation in the use of codes with increasing emergency level. The 20 most frequent ICPC-2 RFE codes accounted for 41% of the green cases compared to 45% of the yellow and 64% of the red. Fever (A03) was the most common green RFE, followed by cough (R05) and abdominal pain (D01). In the yellow urgency level, abdominal pain (D01) and laceration/cut (S18) dominated. Dyspnea (R02) was the third and fever (A03) the fourth most common yellow code. Among the red cases, pain in the chest was the most frequent RFE; A11 (unspecified chest pain) and K01 (heart pain) were by far the two most common red codes, followed by shortness of breath/dyspnea (R02). There were also more neurological ICPC-2 RFE codes (including K90 - stroke/cerebrovascular accident) on the red urgency level top-20 list. Notably, abdominal pain (D01) was the 7th and fever (A03) the 20th most frequent red ICPC-2 RFE code.
Musculoskeletal and general and unspecified issues were the most frequent RFEs in this study. Chapter L and chapter A combined constituted 32.4% of all RFEs, or 110 OOH patients per 1 000 person-years. Respiratory, skin and digestive were also common. General symptoms, respiratory and lower limb issues in addition to fever were the most frequent RFE among the non-urgent (green) cases, and abdominal pain (D01) in the urgent (yellow) ones. Among potentially life threatening cases chest pain dominated. In red urgency level, 25.3% of patients had chest pain related ICPC-2 RFE codes. There was less variation in the use of ICPC-2 with increasing urgency level.
This is the first published study on RFE in Norwegian OOH emergency primary health care. Our institution publishes yearly reports based on all reimbursement claims from doctors in Norwegian OOH services. These claims contain ICPC-2 diagnoses . The three most frequent chapters in reimbursement claims in 2014 were Chapter A (general and unspecified, 24.1%), Chapter R (respiratory, 15.2%) and Chapter L (musculoskeletal, 13.9%), in line with the results of this study. The most frequent ICPC-2 codes in the reimbursement statistics were A99 (general disease, not otherwise specified), R74 (upper respiratory infection, acute), U71 (cystitis/urinary infection) and D01 (abdominal pain). The ICPC-2 frequencies in these reimbursement-based statistics are not directly comparable to the results of this study; mainly because the codes are diagnoses codes set by a doctor, not RFEs set by nurse. The coding might also be biased or less precise (more unspecified codes) because the ICPC-2 code is a prerequisite for the reimbursement, which means that the coding is partly motivated by financial incentives. There is also no information about urgency level in the reimbursement claims.
In a recent study on telephone contacts to Danish OOH primary care , fever was the most frequent RFE. Direct comparison with our study is challenging, since no incidences were presented. Fever constituted a higher proportion of all contacts (10%) than in our results (5.4% of green contacts). In addition, general stomach pain, cough, ear pain and throat symptoms were among the top five Danish RFEs, these were also among the most frequent in our material.
In a study from Switzerland comparing walk-in patients seeking services at an emergency department with a OOH general practitioner cooperative, the frequencies of musculoskeletal (L), general and unspecified (A) and urology (U) were similar to our findings . Respiratory (R), digestive (D) and psychology (P) chapters were considerably more frequent and skin (S) less frequent in the Swiss GP OOH service compared to our findings. Some of the differences could be due to better direct availability of other alternative emergency health care services in the less gatekeeping Swiss system.
ICPC-2 RFE codes have been analyzed in daytime outpatients in general internal medicine department of a small community hospital in Japan . Here, there was less variation in ICPC-2 RFE codes. The 20 most common RFEs accounted for 74% of all cases in the Japanese hospital, considerably more than for all urgency levels in our material. Similar to our findings, R05 (cough) and A03 (fever) were the most common RFEs.
In a study of the epidemiology of medical emergency contacts outside hospitals in Norway , ICPC-2 RFEs were retrospectively assigned to cases classified as potentially life threatening. Similar to red urgency level in this study, cardiovascular events were most frequent, with an incidence of 680 per 100 000 person-years. The 2.5-fold higher incidence compared to our results, can be attributed to the fact that this study was based on data from emergency medical dispatch centers (EMCC), which are not a part of the primary care. Both respiratory and digestive issues were 2.4 times more frequent in this EMCC study compared to our OOH-data.
The main strength of this study is the high number of observations. Also, there were no competing OOH services to the casualty clinics in the investigated areas, which means that we in principle we captured all OOH activity. All relevant cases should thus have been recorded, and this means that the observed incidence rates are trustworthy. RFEs have not previously been systematically recorded in Norwegian OOH services. From previous reports, we know that the quality of the Watchtower data is good. In 2014 only 1.7% of records were missing at least one piece of information. Unavoidably, some of the contacts at the Watchtower clinics are not captured, mostly during periods of high workload. It is not possible to estimate the extent of missing records in our system. This is a potential source of bias, but we believe this is a minor problem since missing cases in the Watchtower project in general are relatively infrequent (estimated 15%), and since our findings are in line with RFE frequencies in other studies. High workload and delay due to technical problems in the web-based ICPC-2 module probably also contributed to isolated missing ICPC-2 RFE codes. Despite missing information, we believe that the ICPC-2 frequencies are representative and with insignificant bias. Less than one in ten records were without information on RFE, and the differences between records with and without ICPC-2 RFE codes were negligible.
We observed large deviations in RFE incidences between individual OOH clinics. Divergent ICPC-2 RFE coding practices and local differences in the organization of emergency health services are possible explanations. The variation may also reflect significant differences in OOH utilization, which we previously have shown are associated with differences in travel distance .
We find it important to stress that we have recorded RFEs, not diagnoses. The recorded ICPC-2 RFE codes are based exclusively on what the patients present upon first contact, they are not the result of an examination by a doctor. The RFEs do therefore not necessarily reflect the actual morbidity of the population. RFEs are often less precise than diagnoses. Symptoms, signs and general contact reasons dominate, and such RFEs provide a good basis for triage and limit bias in the following diagnostic process. The Reason-for-Encounter mode of the ICPC has proven reliable, adequate, and feasible .
Originally, the Watchtowers constituted a representative sample of the Norwegian OOH services . It could be questioned whether results from the projects still are generalizable in 2014 and 2015, but we believe it gives a good picture of the situation in Norway. The Norwegian OOH services (“legevakt”) system has important characteristics that differ from OOH services in most other countries . The compulsory gatekeeping function is one factor that may contribute to significant differences, especially among the most urgent cases.
The results add important knowledge about what frequencies of different issues the staff at Norwegian OOH clinics can expect. The presented incidences may be useful in the planning of how future OOH clinics should be staffed and equipped, and give important clues to which topics that should be given priority in training of nurses and other staff. OOH services are especially suited to handle cases with yellow emergency level, and the high frequency of abdominal pain in this group indicate that this should be emphasized in the education of OOH staff. There is potential for more research on RFEs in Norwegian OOH services, for example in different subgroups of patients. In addition, it would be interesting to compare the incidences in this study with other parts of the health care system, like hospitals and ambulances.
Musculoskeletal, respiratory, skin, digestive and general and unspecified issues were the most frequent RFEs in this study on Norwegian OOH services. Fever was the most frequent single RFE. Abdominal pain was the most common RFE with yellow urgency level, and chest pain dominated the potentially life threatening cases.
This study identifies important differences in RFEs between urgency levels in the Norwegian OOH services. The findings provide new insight into the function of the primary health care emergency services in the Norwegian health care system, and should have implications for training, equipment and staffing of the OOH services.
International classification of primary care
Out of hours
Reason for encounter
Langhelle A, Lossius HM, Silfvast T, Björnsson HM, Lippert FK, Ersson A, Søreide E. International EMS systems: the Nordic countries. Resuscitation. 2004;61:9–21.
Nieber T, Hansen EH, Bondevik GT, Hunskar S, Blinkenberg J, Thesen J, Zakariassen E. Organisering av legevakt [Organization of Norwegian out-of-hours primary health care services]. Tidsskr Nor Laegeforen. 2007;127:1335–8.
Eikeland OJ, Raknes G, Hunskår S. Vakttårnprosjektet. Epidemiologiske data frå legevakt. Samlerapport for 2014. Rapport nr. 3- 2015. [The Watchtower project. Epidemiological data from OOH services 2014]. Bergen: National Centre for Emergency Primary Health Care, Uni Research Health; 2015.
Sandvik H, Hunskaar S. Årsstatistikk fra legevakt 2014. Rapport nr. 2-2015 [Annual statistics from out-of-hours services 2014]. Bergen: National Centre for Emergency Primary Health Care, Uni Research Health; 2015.
Moth G, Huibers L, Christensen MB, Vedsted P. Out-of-hours primary care: a population-based study of the diagnostic scope of telephone contacts. Fam Pract. 2016;33:504–9.
Huber CA, Rosemann T, Zoller M, Eichler K, Senn O. Out-of-hours demand in primary care: frequency, mode of contact and reasons for encounter in Switzerland. J Eval Clin Pract. 2011;17:174–9.
Hansen EH, Hunskaar S. Development, implementation, and pilot study of a sentinel network (“The Watchtowers”) for monitoring emergency primary health care activity in Norway. BMC Health Serv Res. 2008;8:62.
Zoho Corporation. Creator. https://www.zoho.com/creator/. Accessed 18.03.2016.
Wonca International Classification Committee. ICPC for registrering av kontaktårsak [ICPC for recording reason for encounter]. In: ICPC-2. http://www.kith.no/sokeverktoy/icpc-2/bok/kontaktarsak.html. Accessed 18 Mar 2016.
Norwegian Medical Associatrion. Norsk indeks for medisinsk nødhjelp Norwegian Index of Emergency Medical Assistance. 3rd ed. Stavanger: The Laerdal Foundation for Acute Medicine; 2009.
Takeshima T, Kumada M, Mise J, Ishikawa Y, Yoshizawa H, Nakamura T, Okayama M, Kajii E. Reasons for encounter and diagnoses of new outpatients at a small community hospital in Japan: an observational study. Int J Gen Med. 2014;7:259.
Zakariassen E, Burman RA, Hunskaar S. The epidemiology of medical emergency contacts outside hospitals in Norway--a prospective population based study. Scand J Trauma Resusc Emerg Med. 2010;18:9.
Raknes G, Hansen EH, Hunskaar S. Distance and utilisation of out-of-hours services in a Norwegian urban/rural district: an ecological study. BMC Health Serv Res. 2013;13:222.
van der Horst F, Metsemakers J, Vissers F, Saenger G, de Geus C. The Reason-for-Encounter mode of the ICPC: reliable, adequate, and feasible. Scand J Prim Health Care. 1989;7:99–103.
We would like to thank Vivian Midtbø for her work on grouping the ICPC-2 codes, Ole Johan Eikeland for contributions to the data collection, Valborg Baste for statistics advice and above all the staff at the casualty clinics for performing the Watchtower data collection.
The study was funded by National Centre for Emergency Primary Health Care, Uni Research Health, Bergen, Norway.
Availability of data and materials
Data that underlies this study will not be made publicly available. The approval from neither the ethics committee nor from the privacy ombudsman for research allow for disclosure of raw data.
GR contributed to the conception of the study, designed the study, collected data, performed statistical analyses and drafted the manuscript. SH contributed to the conception and design of the study. Both authors participated in revising the manuscript and approving the final version.
Guttorm Raknes (b. 1974) is MD, PhD (dr. philos) and researcher at the National Centre for Emergency Primary Health Care, Uni Research Health, Bergen, Norway. He is approved specialist in clinical pharmacology and works as a consultant at Regional Medicines and Information and Pharmacovigilance Centre (RELIS) in Tromsø, Norway. He has worked for several years as general practitioner.
Steinar Hunskaar (b. 1956) is MD, PhD, professor at University of Bergen, and research director at the National Centre for Primary Health Care. He is also a general practitioner and an approved specialist in general practice.
The authors declare that they have no competing interests.
Consent for publication
Ethics approval and consent to participate
This study is a part of the Watchtower project which is approved by the Regional Ethics committee for Medical Research and by the Norwegian Social Science Data services (privacy ombudsman for research). The Watchtower project database is owned and managed by our institution, the National Centre for Emergency Primary Health Care. The database is not publicly and freely available. These data are anonymized, patient identity is not recorded at any time. Consent from individual patients to participate is not necessary.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Reasons for encounter, ICPC-2 chapters, all urgency levels. Counts, proportions. Total, minimum and maximum incidence in individual OOH casualty clinics. (PDF 79 kb)
Reasons for encounter, ICPC-2 chapters, green urgency level (not urgent). Counts, proportions. Total, minimum and maximum incidence in individual OOH casualty clinics. (PDF 80 kb)
Reasons for encounter, ICPC-2 chapters, yellow urgency level (urgent). Counts, proportions. Total, minimum and maximum incidence in individual OOH casualty clinics. (PDF 80 kb)
Reasons for encounter, ICPC-2 chapters, red urgency level (life threatening). Counts, proportions. Total, minimum and maximum incidence in individual OOH casualty clinics. (PDF 79 kb)
About this article
- After-hours care
- Emergency medical services
- Health services
- Needs and demand
- International Classification of Primary Care (ICPC)
- Primary health care
- Reason for encounter