987
Views
35
CrossRef citations to date
0
Altmetric
LEADING ARTICLE

The Continuous Morbidity Registration Nijmegen: Background and history of a Dutch general practice database

Pages 5-12 | Published online: 11 Jul 2009

Introduction

The Continuous Morbidity Registration (CMR) Nijmegen was founded in 1967 by the founding father of Dutch general practice, Frans Huygen. In a practice-based setting, data were collected concerning health problems patients presented to their general practitioner (GP)—information that was at that time virtually unknown. Huygen was involved in the recording of four practices: in inner-city Nijmegen, in the agricultural villages of Lent and Oosterhout, and in the market town of Doesburg. What started in 1967, never stopped: GPs changed, patients were born, moved into or out of the practice, or died, but the meticulous recording of the health problems presented went on and on. The recording took place in the context of the Dutch healthcare system Citation[1], where patients are listed with a GP for many years and receive all the professional healthcare they need through this GP. As patients commonly join a practice list as a whole family or household, the GP not only serves as a personal doctor but also as a family physician. Data on individual health problems can therefore be related to individual, family, and population demographic characteristics Citation[2], Citation[3].

This issue commemorates the 40-year anniversary of the CMR and presents the core of the experience gained: an overview of health problems and morbidity encountered in general practice over this period. Emphasis, in this overview, is on the dynamics of this experience, on the health problems that have made their mark or disappeared during that period, and on the insights gained by observations over time of the development of individual morbidity. Family patterns and health risks of vulnerable groups have always been of interest, as have psychosocial problems, and this interest will also be apparent. In 1986, a new function of the database was developed: the recording of the process and outcome of care (the Nijmegen Monitoring Project; NMP) Citation[4]. This strengthened the relation between data collection and the care of patients—research and practice—as these data made it possible to monitor and audit care, and provide feedback. This started with the care of patients with diabetes mellitus Citation[4] and has been extended, step by step, to other common chronic diseases. Five other practices joined for this part of the data collection, and, since 1986, the CMR/NMP functions more like a practice-based research network (PBRN) in the North American and UK sense Citation[1]. Summary findings of this experience are also presented in this issue.

History and methodology of the recording and coding of data

An anecdote connected to the founding of the CMR is worth recounting here. At about the same time, Frans Huygen was to be appointed as the inaugural chair of general practice at the medical faculty of the university, now the Radboud University Nijmegen Medical Centre. As one of his first lectures for medical students, he had chosen “German measles”—in that pre-vaccination era, a disease with profound impact on the childhood practice population. No sooner had he set off from his department to the lecture hall—so the anecdote goes—when an official complaint had been tabled by the head of paediatrics, to the dean, against general practice: “measles, as was general knowledge, was a disease of children, and there could be no discussion that childhood morbidity was his—paediatrics—domain”. Such was the lack of empirical information from general practice that even a condition such as measles, hardly ever referred by GPs to paediatricians, could take centre stage in a turf battle at the medical school. If not the trigger to start the CMR, this experience made Frans Huygen even more determined to collect core data concerning the day-to-day experience of GPs.

The challenges of the four GPs in the early years in shaping the CMR must have substantial. There was no classification system available that genuinely covered the health problems presented in general practice. The practices had a combined list of about 12 000 patients, but there was no experience in handling such a large database. Theirs was a fundamental learning-by-doing experience. And, of the decisions they took, a surprising number have become state-of-the-art and characterize current, 2008, practice.

First, they decided to build the database on a population basis—the practice population—and establish the socio-demographic characteristics of that population. Each patient and each family that joined the practice was recorded, irrespective of whether they consulted the GP (). For each individual, sex, age, marital status (single, married/living together, widowhood), and social class were established. Social class allocation has been assigned according to the indicators of CBS Statistics Netherlands Citation[5]: a composite score of education and occupation level, grouped as “low”, “middle”, or “high”. Family composition has been defined as any household acting as a societal unity, regardless of the traditional bonds of marriage. Within each family/household, an individual's position has been marked (partners, children ranked according to their birth order). This has been a trademark of the CMR and has allowed the study of morbidity patterns within families and between generations Citation[6], Citation[7].

Figure 1.  Current socio-demographic characteristics of practice population (CMR) compared to Dutch population (CBS).

Figure 1.  Current socio-demographic characteristics of practice population (CMR) compared to Dutch population (CBS).

The second critical decision was to relate presented morbidity to each individual patient. Each patient on the practice list receives a unique identification number to relate presented morbidity to individual characteristics, including previous morbidity recorded. This makes it possible to relate morbidity data to a population and stresses the fact that general practice, in its community orientation, presents the link between public health and individual primary care. The unique patient number makes it possible to deal with the issue of confidentiality: all morbidity data are recorded in relation to this unique patient number and are sent from the practice to the central database. Only in the practice is it possible to identify and relate data to the individual. All patients on the practice lists have been informed of the practices’ involvement in the CMR and the consequent implications of their morbidity data being recorded and entered into the database and used for research, and all patients have been asked to consent to this. The same principle was alluded to when the recording of the process and outcome of chronic diseases (NMP) was initiated. With very few exceptions, all patients consented to this. In case research of database material requires the collection of additional information from the patient, informed consent has to be collected from the patient through the practice.

The third and most contentious issue in 1967 was how to record morbidity data in general practice. At the time, only the E-book existed Citation[8], and this had been translated for the first Dutch national study Citation[9]. Based on this experience, the CMR recorded morbidity according to this classification framework, and, in order to preserve continuity over time, adhered to this classification.

The first morbidity recordings in general practice triggered a more in-depth classification of health problems in primary care, and, in that respect, development by the World Organization of Family Doctors (Wonca) of the International Classification of Health Problems in Primary care (ICHPPC Citation[10]) followed as a result of the CMR and other databases. The ICHPPC diagnostic definitions have been adopted into the CMR and the NMP, but not the classification as such. Nor was this the case when Wonca developed in 1987 the International Classification of Primary Care (ICPC Citation[11]). Staying with the same classification over time has allowed for longitudinal research into the development of primary care morbidity and individuals’ and families’ medical life histories Citation[7], Citation[12–14].

The NMP monitoring of chronic diseases required the development of additional methodology and structure. Criteria for the process and pursued outcome of care had to be defined. These were derived from the prevailing Dutch College of General Practitioners guidelines Citation[15] that serve as the national reference. For the network, these processes and outcomes are the minimum criteria, and additional criteria have been used for evaluation. In a number of cases, these additions have been included in later reviews of the Dutch College guidelines. In patients’ medical records, a “monitoring file” was structured in which all relevant data were placed. Initially, this was done manually Citation[16], but is now included in electronic medical records Citation[4].

The process and outcome data of each patient are brought together in a central database at the Department of General Practice and structured according to patient and practice. From this database, feedback reports are made that compare actual performance with the guideline criteria as an external reference. This way, practices are compared in terms of their performance, and performance over time is monitored.

In response to the challenges of recording and coding morbidity, the GPs decided to meet regularly to review and compare their classification. From this followed an increasing number of definitions and rules the group had agreed upon. This process is still followed today and has been extended to the monitoring of process and outcome of chronic diseases: the GPs meet monthly, and new GPs who join the practice are trained in recording techniques. The effects are clearly visible in terms of the quality of care and the quality of recorded data: compared to external diagnostic criteria, the validity of the recorded diagnoses has been proven to be well above 80% in a number of studies on a variety of conditions Citation[17], Citation[18]. In an increasing number of patients, their care resembles guideline criteria Citation[4].

This principle of meeting regularly has now been extended to practice assistants. Their contribution to the database is vital—from their check of whether the coding of episodes presented to the practice is complete, to listing new patients in the practices, establishing their demographic characteristics (in particular, coding social class), and seeking informed consent for data recording.

Research in the practice population: Landmark research

Implicitly and explicitly, the CMR/NMP database gets its scientific context from previous and parallel research of the same population. This more or less continuous engagement in research on the practice population attests to the value of the database. With hindsight, five landmark studies stand out. The first brings us back to the founding father, Frans Huygen. In his early years in practice, around the end of the Second World War (1945–1947), epidemics of pneumonia struck the practice population. Malnutrition and other social effects related to the war had increased the risk status of the population, and infectious diseases flourished. Yet, at that time, lobar pneumonia as an infectious disease was poorly understood. Based on his observations in Lent and Oosterhout, Frans Huygen observed their spread through his practice population, analyzing and presenting these observations in his MD/PhD, “Lobar pneumonia as an infectious disease”, which he defended in 1948 Citation[19]. The second study that shaped the CMR was Frans Huygen's observations of morbidity patterns in families, his magnum opus and legacy to family medicine, “Family medicine, the medical life histories of Dutch families” Citation[6].

The third study was the founding study of the monitoring of the outcome of chronic diseases in the database (NMP): the screening of the population of six general practices for cardiovascular risk status, studying the effects of intervention and monitoring long-term outcome Citation[20], Citation[21].

The fourth study was a study of the outcome of diabetes and the effects of its treatment Citation[4], Citation[13]. After establishing the increased premature mortality and (cardiovascular) morbidity in patients with diabetes mellitus in the general practice population, this study was directed at a critical appraisal of the outcome of care. In fact, this study piloted the audit and feedback that subsequently became the model of the NMP.

The fifth study focused on the experienced morbidity in the CMR practice population and related experienced morbidity to what was presented to the GP Citation[22]. From this study, the “Dutch” ecology of medical care Citation[23], Citation[24] can be constructed, confirming that patients only present 10% of the episodes of ill health they experience. Of this, 90% is managed entirely in general practice (Figure 2).

Figure 2.  The ecology of medical care, based on experience of the CMR population.

Figure 2.  The ecology of medical care, based on experience of the CMR population.

Summarizing the 1967–2007 experience

This issue presents an overview of health problems and morbidity encountered in general practice over the period 1967–2007, and presents a summary of the methodology applied and the basic data collected. It is to serve as a documentation of primary care morbidity and the a priori health risks of the general practice population. The presented experience is based on 40 years of the recording and coding of health problems that patients have presented to their GP, but also experience of 40 years of research on the data: cross-sectional and longitudinal analyses from an expanding database in a dynamic population. What was true for the methodology of recording and classifying data was also true for the development of the study methodology. Historic cohort analyses and case-control studies were developed in their scientific rigor, while performed. This has resulted over the years in a substantial number of publications. lists all research papers based on the database that have been published in international peer-reviewed journals.

Table I.  International papers based on CMR/NMP data published in peer-reviewed journals.

The data collection, the audit and feedback, and the studies that used the recorded data would not have been possible without the ongoing commitment of the practices and all their team members—GPs, practice assistants, practice managers and supervisors, and others ( lists their names). This issue is a tribute to their invaluable contribution. In particular, the issue is an acknowledgement of the vital contribution of the statistician who developed and supervised this for the better part of the 1967–2007 period: Henk van den Hoogen has been at the scientific helm of the database since 1975, until his retirement in December 2007, and has made a unique contribution to general practice and primary care. This issue is written by Nijmegen staff members who had the privilege of receiving Henk's friendly supervision, wise council, and sound advice over all these years, to pay their tribute to an unsung hero of international general practice research. Henk, thank you for all of this!

Table II.  Current team members of the practices that made data collection possible.

References

  • van Weel C, Smith H, Beasley JW. Family practice research networks. Experience from three countries. J Fam Pract 2000; 49: 938–43
  • van Weel C. Longitudinal research and data collection in primary care. Ann Fam Med 2005;Suppl 3:S46–51.
  • van Weel C, de Grauw WJC. Family practices registration networks contributed to primary care research. J Clin Epidemiol 2006; 59: 779–83
  • de Grauw WJC, van Gerwen WHEM, van de Lisdonk EH, van den Hoogen HJM, van den Bosch WJHM, van Weel C. Outcomes of audit-enhanced monitoring of patients with type 2 diabetes. J Fam Pract 2002; 51: 459–64
  • CBS (Statistics Netherlands). SBC (standard classification of occupation), 1992/2001 [ in Dutch].
  • Huygen FJA. Family medicine, the medical life history of Dutch families. Dekker and van de Vegt, Nijmegen 1978
  • van den Bosch WJHM, van den Hoogen HJM, Huygen FJA, van Weel C. Morbidity from childhood to adulthood. The medical life history of 103 patients. Fam Pract 1992; 9: 290–4
  • E-book College of General Practitioners. A classification of disease. J Coll Gen Pract (Br J Gen Pract) 1959;2:140–59.
  • Olieman AP. Morbidity in general practice [in Dutch]. Thesis. Leiden: Stenfert Kroese; 1969.
  • Anonymous. ICHPPC-2 defined. Inclusion criteria for the use of the rubrics of the International Classification of Health Problems in Primary Care. Oxford: Oxford University Press; 1983.
  • ICPC Wonca International Classification Committee . International classification of primary care, ICPC-2. Second edition. Oxford: Oxford University Press; 1998.
  • van den Bosch WJHM, Huygen FJA, van den Hoogen HJM, van Weel C. Forty years family-based morbidity data in general practice. Scand J Prim Health Care 1993; 11: 31–2
  • de Grauw W, van de Lisdonk EH, van den Hoogen HJM, van Weel C. Cardiovascular morbidity and mortality of type 2 diabetes patients. Diabetic Med 1995; 12: 117–22
  • van Weel-Baumgarten EM, van den Bosch WJHM, van den Hoogen HJM, Zitman FG. Ten year follow-up of depression after diagnosis in general practice. Br J Gen Pract 1998; 48: 1643–6
  • Dutch College of General Practitioners. Guidelines Programme. http://nhg.artsennet.nl (accessed October 4th, 2008). See also for English: http://nhg.artsennet.nl/content/resources/AMGATE_6059_104_TICH_L93909649/AMGATE_6059_104_TICH_R1199541362500411// ( accessed October 4th, 2008).
  • Grauw de WJC, Lisdonk van de EH, Hoogen van den HJM, Weel van C. Monitoring of NIDDM in general practice. Diabetes Nutr Metab;4(Suppl):55s–64s.
  • van Weel C. Validating long term morbidity recording. J Epidemiol Commun Health 1995; 49(Suppl 1)29–32
  • van Weel-Baumgarten EM, van den Bosch WJHM, van den Hoogen HJM, Zitman FG. The validity of the diagnosis of depression in general practice: is using criteria for diagnosis as a routine the answer?. Br J Gen Pract 2000; 50: 284–7
  • Huygen FJA. Lobaire pneumonie als infectieziekte [Lobar pneumonia as infectious disease]. Dissertation. University of Utrecht, Utrecht 1948
  • Bakx JC, Veldstra IM, van den Hoogen HJM, Zielhuis GA, Thien T, van Weel C, et al. Blood pressure and cardiovascular morbidity and mortality in a Dutch population. The Nijmegen Cohort Study. Prev Med 2001; 32: 142–7
  • van de Lisdonk EH. Perceived and presented morbidity in general practice. Scand J Prim Health Care 1989; 7: 73–8
  • van Ree JW. The Nijmegen Intervention Project. Thesis [in Dutch]. University of Nijmegen, Nijmegen 1981
  • White KL, Williams TF, Greenberg BG. The ecology of medical care. N Engl J Med 1961; 265: 885–92
  • Green LA, Fryer GE, Yawn BP, Lanier D, Dovey SM. The ecology of medical care revisited. N Engl J Med 2001; 344: 2021–5

Reprints and Corporate Permissions

Please note: Selecting permissions does not provide access to the full text of the article, please see our help page How do I view content?

To request a reprint or corporate permissions for this article, please click on the relevant link below:

Academic Permissions

Please note: Selecting permissions does not provide access to the full text of the article, please see our help page How do I view content?

Obtain permissions instantly via Rightslink by clicking on the button below:

If you are unable to obtain permissions via Rightslink, please complete and submit this Permissions form. For more information, please visit our Permissions help page.