Abstract
Objective. To estimate the frequency of psychological and social classification codes employed by general practitioners (GPs) and to explore the extent to which GPs ascribed health problems to biomedical, psychological, or social factors. Design. A cross-sectional survey based on questionnaire data from GPs. Setting. Danish primary care. Subjects. 387 GPs and their face-to-face contacts with 5543 patients. Main outcome measures. GPs registered consecutive patients on registration forms including reason for encounter, diagnostic classification of main problem, and a GP assessment of biomedical, psychological, and social factors’ influence on the contact. Results. The GP-stated reasons for encounter largely overlapped with their classification of the managed problem. Using the International Classification of Primary Care (ICPC-2-R), GPs classified 600 (11%) patients with psychological problems and 30 (0.5%) with social problems. Both codes for problems/complaints and specific disorders were used as the GP's diagnostic classification of the main problem. Two problems (depression and acute stress reaction/adjustment disorder) accounted for 51% of all psychological classifications made. GPs generally emphasized biomedical aspects of the contacts. Psychological aspects were given greater importance in follow-up consultations than in first-episode consultations, whereas social factors were rarely seen as essential to the consultation. Conclusion. Psychological problems are frequently seen and managed in primary care and most are classified within a few diagnostic categories. Social matters are rarely considered or classified.
Knowledge about general practitioners’ (GPs’) assessment and classification of psychological and social problems in routine care is limited.
GPs managed psychological problems as the main problem in 11% of their patient contacts; depression and acute stress reaction/adjustment disorder accounted for half of these.
Only 18 of 43 psychological codes in the mental health chapter of the International Classification of Primary Care were applied in more than one per one thousand patients.
Social problems were classified in only 0.5% of the patients and social matters were rarely seen as important according to the GPs’ assessments of contributing factors.
Introduction
Our knowledge concerning general practitioners’ (GPs’) classification of psychological and social problems in routine care is limited and there is a need for empirical studies of classification in primary care to guide quality improvement in clinical practice and the ongoing revisions of the international classification systems of mental disorders [Citation1–4].
Classification of problems and disorders differs significantly between primary and secondary care. Primary care manages starting-point problems and sub-threshold disorders as well as specific diseases, whereas secondary care classifications deal with end-point diagnoses [Citation5,Citation6].
The International Classification of Primary Care (ICPC) included in the WHO family of international classifications has been widely implemented in primary care [Citation7,Citation8]. In order to be inclusive and to make classification categories applicable to all primary care settings, the ICPC minimizes the number of categories (currently 686 rubrics) and gives the same emphasis to classification of problems/complaints as to specific disorders [Citation6,Citation9]. According to the principles of the ICPC, a rubric is assigned when a disorder has a prevalence of at least one per thousand.
However, classification systems also need to be sufficiently specific to guide optimal treatment and quality improvement initiatives. Furthermore, patient care may involve specialized care, and precise communication of diagnoses across sectorial interfaces is essential. These issues raise an important question of the prevalence and clinical importance of classification categories applied in primary and secondary care.
Aims of the study
The study aimed to estimate the frequency of psychosocial problems as the patient's reason for encounter and as the concluding diagnostic categories employed by GPs. Furthermore, the study explored the extent to which GPs ascribed health problems to biomedical, psychological, or social factors.
Material and methods
The present cross-sectional study draws its data from a survey of the activity in Danish general practice conducted from December 2008 to December 2009 [Citation10]. All 871 GPs in the Central Denmark Region (covering approximately 20% of the entire Danish population) were invited to participate. Participants registered all patient contacts during one randomly assigned day. The GPs received payment for their participation (€32) and for each registered contact (€3 per contact). Information about GPs and listed patients was obtained from the Danish National Health Service Register [Citation11].
Each contact was registered on a one-page registration form by the GP. The form encompassed a range of questions including: the personal registration number, gender and age, the type of contact, the main reason for encounter, the diagnostic classification of the main problem, and “As a GP, to what extent do you think the following factors influenced the present contact with the patient? Biomedical __%, psychological __%, social __% or don't know”. The GPs’ answers to each of these dimensions were later dichotomized into major issue (> 50%) or not major issue (< = 50%).
For the purpose of the present paper, we only included GP face-to-face contacts (consultations and home visits) with identifiable patients. We excluded contacts with missing information on type of contact and consultations stated on the registration form as prophylactic only. If patients appeared more than once in the study only their first contact was included ().
Figure 1. Flow diagram of the inclusion of patients.
When reasons for encounter and diagnoses were only stated in text by the GPs they were translated into ICPC-2 codes by an ICPC-trained medical student using an electronic standard terminology (www.dak-e.dk/icpc). All ICPC codes were reviewed by one of the authors (GM).
Although the GPs were asked only to classify the main problem raised in the consultation, some of them stated more than one problem (207 contacts; 3.7%). In these cases, we selected for analysis the first problem entered on the registration form. In 741 cases (13.4%), the main problem was missing and the stated reasons for encounter were extrapolated to replace the missing codes. If the reason for encounter was described as a process (e.g. blood test), the ICPC-2 process codes were applied.
Data management and analysis
Using chi-square, we examined the participating GPs’ representativeness with respect to gender, seniority, type of practice (single-handed or partnership), and listed patients. Reasons for encounters and diagnoses were categorized into symptoms, diseases, and processes according to the ICPC-2-R classification. We used generalized linear models (GLM) with log link and the Bernoulli family [Citation20] to estimate associations between classification (symptom vs. disease codes) and first-episode and follow-up consultations. Likewise, we used GLM to compare first-episode and follow-up consultations for ratios of < = 50% vs. > 50% of the biomedical, psychological, and social factors of the contacts. Multilevel models were applied in order to adjust for patient gender, age, and clustering of patients at the GP level (robust variance estimates).
Each contact registration form was optically scanned using Teleform 8.0. Stata 11.1 was used to perform the statistical analyses.
Ethics
The project was approved by the Danish Data Protection Agency (J.no. 2008-41-2195) and by the National Board of Health (J.no. 7-604-04-2/49/EHE). According to Danish law, approval from the ethical committee was not needed.
Results
A total of 404 GPs (46.4%) consented to participate in the survey. Seventeen (4%) were excluded as they did not state the personal registration number of any of their patients. The 387 participating GPs and the non-participating GPs within the catchment area did not differ with regard to type of practice (p = 0.933), but we found higher representation of female GPs (44.4% vs. 38.9% in the region, p = 0.003) and lower representation of GPs working more than 20 years in practice (20.4 vs. 25.5, p = 0.017). Patients listed with participating GPs were comparable to those listed with non-participating GPs with regard to age and gender distribution (p = 0.354 and 0.389, respectively).
A total of 5543 patients with face-to-face contacts (5357 consultations and 186 home visits) formed the population of the present study (see ).
Reasons for encounter
shows the GP-reported reasons for encounter divided into psychosocial and physical reasons. Overall, 553 patients (10.0%) visited for psychological and 38 patients (0.7%) for social problems. Overlap between reasons for encounter and GPs’ diagnostic classification of main problems was substantial; 518 of 553 patients (93.7%) with psychological reasons for encounter were classified in the psychological chapter, whereas only 82 of 4990 patients (1.6%) with other reasons for encounter were classified by the GP as psychological.
Table I. Reasons for encounter and classification of patients with face-to-face contacts (n = 5543) reported by general practitioners.
Diagnostic classification of main problem
GPs classified 600 patients (10.8%) in the psychological chapter; 392 (65.3%) came for follow-up consultations (). Stratification revealed that psychological codes were applied to 33 children < 18 years (3.9% of the children), to 364 female adults (12.5% of the women), and to 203 male adults (11.4% of the men). Both symptom codes and specific codes were used widely as diagnostic classification for psychological problems, although symptom codes were fewer at follow-up consultations than in first-episode consultations (p < 0.001).
Table II. Distribution of psychological problems and diagnoses (ICPC Chapter P).
Two rubrics accounted for 51.3% of all psychological codes (P76 Depression (39.8%) and P02 Acute stress reaction/adjustment disorder (11.5%)) (). The distribution of these codes did not differ between men and women (p = 0.475). Children had a different pattern; three rubrics comprised 54.6% of all psychological codes: bedwetting/enuresis (21.2%), hyperkinetic disorder (18.2%), and acute stress reaction/adjustment disorder (15.2%). Ten rubrics were not used at all.
Table III. Frequency of applied rubrics for GP classification of psychological problems.
Only 30 patients (0.5%) were classified in the social chapter: three children (0.4%), 21 female adults (0.7%), and six male adults (0.3%). Five codes covered 70% of these patients: Work problem (6), Social problem NOS (5), Social cultural problem (4), Loss/death of partner problem (3), and Relationship problem parent/family (3).
GPs’ emphasis on psychosocial aspects
The GPs’ assessments of biomedical, psychological, and social factors independent of their diagnostic classification were analysed for each factor separately. GPs weighted biomedical aspects higher than psychosocial aspects. In 2798 patients (52.0%), biomedical factors were seen as the sole contributors, whereas only 304 patient contacts (5.6%) had no biomedical aspect according to the GP. The mean values for the three factors were as follows: biomedical 78.0% (95% CI 77.2–78.9), psychological 17.1% (95% CI 16.3–17.8), and social 4.7% (95% CI 4.4–5.0).
Among the 304 patients in whom the GPs found that no biomedical aspects were involved, 227 (74.7%) were classified in the psychological chapter and 22 (7.2%) in the social chapter of ICPC.
shows the dichotomized assessments by type of episode. Excluding missing items from this table in statistical analyses showed that GPs gave more emphasis to biomedical aspects in first-episode consultations than in follow-up consultations (factor > 50% in 82.6 vs. 67.8% of cases, p < 0.001), whereas more emphasis was ascribed to psychological aspects in follow-up consultations than in first-episode consultations (factor > 50% in 6.0 vs. 13.6% of cases, p < 0.001). We found no statistically significant difference for social factors (p = 0.348).
Table IV. Distribution of biomedical, psychological, and social factors contributing to the main problem of patient contacts according to the general practitioner.
Discussion
Statement of principal findings
The GP-reported reasons for encounter and their diagnostic classification of main problems were largely overlapping with regard to psychological problems. GPs classified psychological problems in 11% of their patient contacts. Of 43 ICPC rubrics from the psychological chapter, only 18 occurred in more than one per one thousand patients and two rubrics (depression and acute stress reaction/adjustment disorder) accounted for half of all the psychological codes employed.
GPs focused primarily on biomedical aspects in all contacts, but even more so in first-episode consultations. With regard to psychological contributions, more emphasis was given in follow-up consultations compared with first episodes.
Social problems were hardly classified at all and were rarely considered a contributing factor.
Strengths and weaknesses of the study
The study was based on the largest and most extensive survey of primary health care encounters in Denmark, on the whole representative of the Danish primary care setting. However, the setting itself is selected as it consists of highly educated GPs who are familiar with the application of diagnostic criteria. We do not know the extent to which the results will apply to other, e.g. non-Western, primary care settings.
GPs were allowed to state only one reason for encounter and one diagnostic label for each patient, but patients in primary care often present several problems in the same consultation. While this should not affect the distribution of rubrics, our prevalence values of psychological and social problems and diagnoses may therefore in general be underestimated. On the other hand, our approach allowed analyses to be based on patient numbers as the denominator and made sure that the problem or diagnosis stated was, in fact, managed by the GP.
The classification itself may also be a limitation. The GPs’ diagnostic labelling by the use of ICPC codes or text was not validated and the GP inter-rater variability is known to be substantial at rubric level [Citation12]. Furthermore, the after-coding of the GPs’ classification text may have introduced some degree of bias as the GPs’ text entries were not always unambiguous. This bias would mainly affect the rubrics within the ICPC chapters and not between the chapters.
Finally, we were not surprised to find much overlap between reasons for encounter and diagnostic classification of main problem as both were stated by the GPs. The results might have been different if analyses had been based on patient statements.
Comparison with existing literature
Our reported frequency of psychological problems is lower than the 25–50% reported in prevalence studies of mental disorders in primary care [Citation13–16]. However, these studies are of primary care waiting-room populations. Our results are comparable to those found in studies of problems managed by the GP in the consultation. An Australian study reported that 11.5 psychological problems were managed per 100 encounters, the most common with frequencies above 0.01 being: depressive disorder (3.8%), anxiety, sleep disturbance, acute stress reaction (0.6%), drug abuse, schizophrenia, dementia, tobacco abuse, alcohol abuse, and affective psychosis [Citation17]. An international study showed comparable results for Dutch and US primary care diagnoses, whereas prevalence estimates were substantially lower for Japan and Poland [Citation18]. Finally, a Norwegian study from 1994 described frequencies of psychological problems and social problems of 7% and 1%, respectively [Citation19]. In agreement with the low prevalence of social problems, another Norwegian study found that patients rarely disclose social problems to their GP and GPs are often unaware of general social contributors to the patients’ health problems [Citation20,Citation21].
Also in psychiatric care, few diagnoses cover the majority of patients cared for. In a recent Danish study, 16 diagnoses accounted for 50% of all contacts, schizophrenia, alcohol abuse, and adjustment disorder being the most frequent [Citation22]. Compared with psychiatric care [Citation22,Citation23] the classification of schizophrenia and alcohol abuse was less common in primary care but reached frequencies above 0.01 which is the limit for being assigned a rubric in the ICPC.
Conclusion and perspectives
Psychological problems and mental disorders are frequent in primary care, but are captured using rather few diagnostic categories. Applying 18 diagnostic categories, all with a frequency above one per thousand, would still enable communication with secondary care on diagnoses frequently applied in that setting. However, the development of future primary care classifications would gain from studies that embrace a broader range of primary care settings and the inclusion of primary care terminologies rather than only the categories of existing classification systems.
Surprisingly, GPs rarely stated social factors as being important to the presented problem. This was apparent not only from their classification, but also from their independent statements regarding the biomedical, psychological, and social contributions to the patients’ problems. GPs generally paid much attention to biomedical aspects and when they dealt with psychological issues this happened more frequently in follow-up consultations. Hence, the bio-psycho-social model was deployed in a highly skewed way and seemed to be applied in a sequential manner across consultations rather than within the confines of the individual consultation. This raises a question not only about the contents and quality of the classification of psychological and social problems, but also about where GPs place their focus in the consultation process and how this affects their classification and management of problems and disorders.
Acknowledgements
The authors would like to thank the GPs and their staff for their contributions to this study.
Declaration of interest
The authors report no conflict of interest. The authors alone are responsible for the content and writing of the paper.
M. Rosendal is a member of Wonca's International Classification Committee and of the Primary Care Consultation Group, International Advisory Group for the Revision of ICD-10 Mental and Behavioural Disorders, the World Health Organization.
The study was funded by “Samarbejdsudvalget for Århus Amt”, “Fonden vedrørende Finansiering af Forskning i Almen Praksis og Sundhedsvæsenet i øvrigt”, the Research Fund for General Practice (“Temapulje: almen praksis i sundhedsvæsenet”), and the Committee for Quality Improvement and Continuing Medical Education in Central Denmark Region. The funding sources had no role in the study and did not influence the data collection, data analyses, or publications.
References
- WHO. Revision of the International Classification of Diseases (ICD). Available at: http://www.whoint/classifications/icd/ICDRevision/en/2011
- Gask L, Klinkman M, Fortes S, Dowrick C. Capturing complexity: The case for a new classification system for mental disorders in primary care. Eur Psychiatry 2008;23:469–76.
- Jenkins R, Goldberg D, Kiima D, Mayeya J, Mayeya P, Mbatia J, . Classification in primary care: Experience with current diagnostic systems. Psychopathology 2002;35:127–31.
- Aboraya A. Scientific Forum on the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-V) – An Invitation. Psychiatry (Edgmont) 2010;7:32–6.
- Rosendal M, Falko E. [Diagnostic classification in Denmark with emphasis on general practice]. Ugeskr Laeger 2009;171:997–1000.
- Lamberts H, Magruder K, Kathol RG, Pincus HA, Okkes I. The classification of mental disorders in primary care: A guide through a difficult terrain. Int J Psychiatry Med 1998;28:159–76.
- Soler JK, Okkes I, Wood M, Lamberts H. The coming of age of ICPC: Celebrating the 21st birthday of the International Classification of Primary Care. Fam Pract 2008;25:312–17.
- De Lusignan S, Minmagh C, Kennedy J, Zeimet M, Bommezijn H, Bryant J. A survey to identify the clinical coding and classification systems currently in use across Europe. Medinfo 2001;10:86–9.
- Wonca. International Classification of Primary Care. ICPC-2-R. 2nd ed. New York: Oxford University Press; 2005.
- Moth G, Olesen F, Vedsted P. Reasons for encounter and disease patterns in Danish primary care: Changes over 16 years (a historical analysis). Scand J Prim Health Care 2012;30:70–5.
- Olivarius NF, Hollnagel H, Krasnik A, Pedersen PA, Thorsen H. The Danish National Health Service Register: A tool for primary health care research. Dan Med Bull 1997;44:449–53.
- Britt H, Angelis M, Harris E. The reliability and validity of doctor-recorded morbidity data in active data collection systems. Scand J Prim Health Care 1998;16:50–5.
- Linden M. Mental disorders in primary care. Adv Psychosom Med 2004;26:52–65.
- Serrano-Blanco A, Palao DJ, Luciano JV, Pinto-Meza A, Lujan L, Fernandez A, . Prevalence of mental disorders in primary care: Results from the diagnosis and treatment of mental disorders in primary care study (DASMAP). Soc Psychiatry Psychiatr Epidemiol 2010;45:201–10.
- Hanel G, Henningsen P, Herzog W, Sauer N, Schaefert R, Szecsenyi J, . Depression, anxiety, and somatoform disorders: Vague or distinct categories in primary care?Results from a large cross-sectional study. J Psychosom Res 2009;67:189–97.
- Toft T, Fink P, Oernboel E, Christensen KS, Frostholm L, Olesen F. Mental disorders in primary care: Prevalence and co-morbidity among disorders. Results from the Functional Illness in Primary care (FIP) study. Psychol Med 2005;35:1175–84.
- Harrison CM, Britt H. The rates and management of psychological problems in Australian general practice. Aust N Z J Psychiatry 2004;38:781–8.
- Okkes IM, Polderman GO, Fryer GE, Yamada T, Bujak M, Oskam SK, . The role of family practice in different health care systems: A comparison of reasons for encounter, diagnoses, and interventions in primary care populations in the Netherlands, Japan, Poland, and the United States. J Fam Pract 2002;51:72–3.
- Claussen B, Famm E, Nygard JF. ICPC-diagnoser i 60 allmennpraksiser [ICPC diagnoses in 60 offices of general practitioners]. Tidsskr Nor Laegeforen 1994;114:821–4.
- Gulbrandsen P, Fugelli P, Hjortdahl P. Psychosocial problems presented by patients with somatic reasons for encounter: Tip of the iceberg?Fam Pract 1998;15:1–8.
- Gulbrandsen P, Fugelli P, Hjortdahl P. General practitioners’ knowledge of their patients’ socioeconomic data and their ability to identify vulnerable groups. Scand J Prim Health Care 1998;16:204–10.
- Munk–Jorgensen P, Najarraq LM, Bertelsen A. Use of ICD-10 diagnoses in Danish psychiatric hospital-based services in 2001–2007. World Psychiatry 2010;9:183–4.
- Mussigbrodt H, Michels R, Malchow CP, Dilling H, Munk-Jorgensen P, Bertelsen A. Use of the ICD-10 classification in psychiatry: An international survey. Psychopathology 2000;33:94–9.