Family Practice Advance Access originally published online on February 27, 2008
Family Practice 2008 25(2):132-136; doi:10.1093/fampra/cmn002
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Research and practice combined—ideas for a life in general practice
Institute of Community Medicine, University of Tromsø, 9037 Tromsø, Norway
Email: knutarne.holtedahl{at}ism.uit.no
Received 20 April 2007; Revised 5 November 2007; Accepted 8 January 2008.
 |
Abstract |
|---|
Top
Abstract
IntroductionAbstract: The modern era of research in general practice is scarcely more than half a century old. The author has been fortunate enough to be part of this development for more than three decades, sharing his professional life between clinical practice and research. Here is the story. All clinical care should be supported and developed by research evidence, and some of it must be collected where the care is performed. Research may improve the theoretical understanding underlying practice. Some patients probably receive better care because their general practitioner has done research, or their GP has attended a meeting or read an article by a research-minded colleague.
Keywords. History of medicine, clinical diagnosis, family medicine, researchers.
|
Introduction |
|---|
Research is a slow activity. It requires reflection, patience and accuracy more than action. Things fall apart, and repair work is always more urgent. For the young doctor I was in 1970, arriving in rural Niger fresh from a Paris half-year course of tropical medicine, learning everyday clinical practice became overwhelmingly important. Still, some observations invited more systematic study. Among all the underweight children coming to the hospital, why were none less than 6 months old? A research question was born, and for a while I made systematic measurements and notes. Then it stopped. I did not know how to put data together and even less about how to analyse them.
Coming back to Norway, general practice quickly appealed to me. Turnover was rapid and the challenges varied. A temporary passage over to a psychiatric therapeutic community made me wonder: When institutional psychiatry has been shown to take care of less than 10% of patients having a psychiatric problem, what does the GP do for the rest? Why not ask the GPs? I created a questionnaire and mailed it to all GPs in my region. I counted the different answers through markings on the edges of a carton board and put the results together in a manuscript. No very original findings: GPs spoke with their patients, and sometimes they prescribed drugs. A kind referee suggested how I could revise the manuscript. By so doing, he became my first teacher of writing articles. I started over on my typewriter, and the journal's acceptance encouraged vigilance for other interesting observations and questions.1
Another question arose in my first real general practice job. Among several patients presenting with adenitis of the neck, one young woman turned out to have a nasal carcinoma, and one woman got repeated treatment for cystitis until her apparent hematuria turned out to come from a cancer in her genitals. Rarities, and perhaps no wonder I did not go straight to the diagnosis. I still felt I could have done better when the truth was unveiled after 3–4 months. At least I was certain that both patients were asking themselves why their GP took so long to understand what they had. How to do better?
|
Looking at what doctors had been doing |
|---|
I wrote some optimistic letters. First, to my GP colleagues and to the University hospital to ask permission to study their medical records, and then to the Cancer Registry of Norway to have a list of all cancer patients diagnosed in 1976. The latter was the most difficult, but permission was granted at last. Ethical committees and data surveillance agencies were not yet thought of. I wanted to find the records—hospital and general practice—for each patient to see what led to diagnosis, how long it took and what I subjectively could conclude about the quality of the medical work. Not to see who were bad doctors, but to understand how things might improve, an argument generously accepted by my colleagues. For the first time, I went to the medical literature to see what others had done in the field. What I found all came from hospital studies. I found the ‘delay’ terms2: patient delay for the patient who waits too long before consulting and doctor's delay when referral or treatment was not adequately done. Rural doctor leave made it possible to spend 1 month with all the journals, in the hospital basement and in the practices. I proposed my own subjective definitions of delay and found that half the patients had delay of one kind and half the patients of the other, with considerable overlap of the groups.3 It was only years later, I discovered Feinstein's articles about what is possible and not possible to read out of a medical journal.4 However, some practical advice was possible to give, it was possible to learn from ordinary mistakes we all had been doing.5 Improvement was possible through a better medical history, a better clinical examination and more focused supplementary examinations.
|
How to study symptoms? |
|---|
This led to a closer look at symptoms. How were symptoms understood by patient and by doctor, and when did they mean what? A combination of newly instituted short-time research grants for GPs and a new Department of general practice at the young university gave me my first feedback on scientific method. Epidemiology could be clinical6 and apply to smaller groups of patients. The validity terms applied to laboratory tests could be applied to symptoms, and a symptom may be present and absent in the same way as a test is positive or negative. Symptoms had a sensitivity, a specificity and therefore also a likelihood ratio. Symptoms are unspecific in relation to cancer or any other well-defined disease, and intuitively more so in general practice than in hospitals—a truth with some limitations, it would turn out. Predictive values vary with prevalence, an essential insight for understanding how clinical thinking and practice are different in general practice and in hospitals. This research is not readily available; my first articles were written in Norwegian, some of my English language articles in journals with referees but no Medline indexing,7,8 and my thesis was published in a University series,9 a few can be found in present-day databases.10,11
|
Technological revolution and institutional change |
|---|
During these years, technology changed dramatically. Computers with large disks replaced typewriters, but text treatment required slow processes of formatting for each new printed version. In the new personal computers, one of the two diskette stations was soon replaced by a hard disk, 20 MB at first. Data registration programs no longer required extensive programming expertise, but recoding and analyses were slow. Dealing with several programs at the same time was a dream, and literature was on paper and took weeks to order. A kind university librarian arranged for me a subscription for a monthly Medline search on ‘general practice and cancer', an access I was not normally allowed, not being employed by the university.
For weeks or months at a time I worked at the Department of General Practice. The department in Tromsø is integrated in the Institute of Community Medicine, dominated at the beginning by epidemiologists, most of them with some experience from clinical medicine including general practice. The institute's main idea was to perform large-size population studies, but it was also responsible for the teaching of public health. The faculty had a clear political challenge to educate doctors for the sparsely populated northern regions, and the Department of general practice was given freedom to introduce practice periods in local primary care as well as theoretical undergraduate teaching seen from the general practice viewpoint. An interaction between teachers from different departments thus developed and has proved fruitful also when one looks at the present-day situation in academic departments of general practice internationally; GPs with quite varied academic competence collaborate with sociologists, anthropologists, epidemiologists, statisticians and philosophers to teach and to explore the very rich research possibilities in primary care. But basically, in Tromsø and elsewhere, the gradually more extensive participation of GPs in undergraduate education has been the backbone for financing academic positions and thus creates possibilities for research and research guidance.
|
Practice or theory—or both? |
|---|
Until 1989, I was still a full-time GP. Two factors made it possible to take breaks for research: as a GP in northern Norway, I was allowed special rural doctor leave periods, and small GP research grants covering up to 6 months pay had been instituted by the Norwegian Medical Association. During these years, I went back to the question of underweight children,12 this time in a Norwegian setting very different from the one in Niger, and children with delayed psychomotor development.13 I became more confident that new ideas continued to come up, and the new academic field of general practice looked more and more as a thrilling challenge. Coming back from some new challenges in Africa and a glimpse into the new challenges of HIV,14 I applied for a university position and told my GP colleagues I would like to work 1 day a week in the surgery and twice a month on evening/night emergency duty. Except for nights on duty, I still work like that.
University work does not mean that one can spend every day designing and carrying out research. Teaching is essential and can also be subject to research.15 It is a challenge to preserve general practice's newly achieved status as a major topic in Norwegian medical school curriculum. Still it is a privilege to have time to reflect. Networking is essential.16 Reading specialized literature in many cases became necessary, but I soon found out that cooperation with hospital specialists avoided inventing the wheel over again. Patient recruitment and much of the follow-up could be performed in general practice, but specialists can perform high-tech follow-up examinations. We studied glaucoma17,18 and urinary incontinence 19–21 this way. Critical reading became important.22
|
‘Softer’ methods? |
|---|
Two seemingly different approaches affected research in and outside general practice from the 1990s: Evidence-based medicine and qualitative methods like phenomenological analysis. Medicine deals with answers to the question ‘what kind of treatment, but general practice is more concerned with "who needs treatment?"’.23 Therefore, and because patient populations differ,24 results from randomized controlled trials are more often useful in hospital medicine than in general practice, although certainly not without interest.21 More problematic has been the sometimes heated discussion about what methods provide ‘evidence'. GPs have been among the researchers who have seen the potential of a sound philosophical theoretical base25,26 and of various methods applied in social sciences.27–30 I see a potential for enriching our findings from clinical epidemiological studies,31 and for acquiring knowledge about topics that cannot be studied by counting or measuring.
|
Primary care research in poor countries |
|---|
It is difficult to distinguish between primary and hospital care in many countries, especially in poor countries where access to doctors are limited to hospitals. In later years, I have made several trips to Cameroon.32 HIV treatment may be a highly specialized field in rich countries, but in poor countries, nurses and inexperienced doctors often are responsible for treatment with antiretroviral drugs. Trapped in poverty, clinical work remains an ad hoc affair, and clinical research even more so. But Africa is a continent with immense possibilities for clinical research.33,34 When you live below the poverty index of 2$ a day, it is fantastic when someone covers your HIV drugs. But what if you still have to pay for your tests, your consultations, your prophylactic drugs and your travelling to the hospital? How can you understand the need for life-long medication, especially if you cannot read or write and the drugs make you sick? To see the organization of care from the viewpoint of public health and primary care is currently for me one of several research topics. HIV being another great imitator disease, it also represents new challenges for early clinical diagnosis. Researchers from primary care are well positioned for this kind of research.
|
What has changed and what is unchanged? |
|---|
Technological tools improve with increasing pace. Working with several different software programs at the same time was a dream that materialized more than a decade ago. The creation and modification of data files have become simple and rapid and no longer requires advanced programming skills. Software for recording of data also performs all kinds of statistical analyses and produces graphs if needed. Data programs for analysis of text have been developed. Recently, we have got software for machine-readable questionnaires and for data collection from questionnaires via Internet. Paper questionnaires undergo visual check when the machine is in doubt. Data files are created automatically and are ready for analyses after minor adjustment. The loss of closeness to data is a price to pay for non-manual procedures, but time saving is impressive. Possibly, systematic error is a greater threat with automated systems, while errors of transcription are less likely to occur.
Improved access to literature is important both for clinical practice and for research. More and more GPs now access free-of-charge databases and electronic versions of journals through their regional university library or their medical association. Major medical journals offer selected or unselected content for their latest issue sent by mail. In many cases, mouse clicks may open an abstract or, if needed, the article text as well as its referenced articles. Access to older articles may be a problem, although studying literature lists of review articles may be useful. Downloading to personal databases has become possible.
Better large-capacity computers and development within electronic transfer of data invite better use of the large amount of clinical data collected in general practice. Such network studies could go far beyond a simple counting of categories. However, it is important to acknowledge the difference between data collected to answer a defined research question, and research questions being defined because data are available. The data collection process will normally be more thorough, and the quality or validity of the data better with the former approach.
Access to funding has improved in some countries. In Norway, general practice research units associated with the existing university departments will be established and financed from 2007. This has been a long process and is not more than a dream in most European countries. To be allowed to compete realistically for funding is a must in a professional world. Politicians must understand how primary and secondary care are complementary not only in clinical practice but also when it comes to producing new knowledge. For countries with no financing of general practice research, it might be a temporary inspiration that the first 20–25 PhD degrees from general practice in Norway were initiated with modest grants lasting a few months and financed by our own medical association. This in turn led to further support from other sources, allowing enough time to complete a thesis and qualify fully for academic work.
Still, some things do not change.35 Research ambitions require curiosity and ability to ask questions based on your everyday experience, professional or non-professional. There are examples of GPs who have contributed important data by following local cohorts of people through most of their career,36 and GPs can be part of a research network or participate in studies at a university institute. Geographical remoteness, seen from the northernmost university in the world, is a relative term and should never discourage research activities. A geographic periphery may have advantages that compensate for a sparse population: willingness of people to participate, with higher participation rates, stability of the population and closeness to people in the survey. A rural doctor sometimes is better equipped than city doctors, and specialists in local hospitals may be more willing to delegate interesting tasks.
|
Is it worth while? |
|---|
Research in primary care is no evident goal for everybody in the medical community,37 but all clinical care should be supported and developed by research evidence, and some of it must be collected where the care is performed.38 Family medicine is a research discipline for three reasons:39 unique epidemiology, important context of care and community links. I think clinical practice should include a reflective attitude, encouraging the generation of questions. This is so for all cultures, rich or poor. For me, questions and research have made my practice more interesting, and I am not the one to judge its impact. But I think my research in some instances may have improved my own theoretical understanding of how to deal with some of my patients and that some may have received a more efficient care this way. When I publish or I participate in seminars or teaching, I share my understanding with other doctors, and now and then there will be a cancer patient who is diagnosed earlier, a woman with urinary incontinence who receives more appropriate examination and treatment or a glaucoma patient whose vision is preserved. Even in poor Cameroon, there may be some HIV patients who receive appropriate care because of the ‘primary care’ research in which I have the privilege to participate. Research in general practice is of interest to individual people and to the community they are members of. It is high time that more countries acknowledge this and create and finance programs accordingly.
A GP who wants to do research should be aware of the ways new knowledge is produced. With the advancement of theoretical biological knowledge, new tailor-made drugs or diagnostic methods are systematically searched, and this is perhaps an underestimated approach among clinical researchers. Screening is a different kind of approach, like when the pharmaceutical industry screens thousands of plants with more or less drug-promising possibilities. However, in clinical practice our tests are imperfect, and the burden of biological ‘noise’ usually outweighs the potential gains. Most important is the recognition that great new discoveries most often come as a surprise for mainstream science. Nobel winning topics like prions or helicobacter ulcers met with indignation and objections when the theories were introduced and most of the great technological advances of the 20th century were foreseen by nobody at the beginning of that century. There is no recipe for original thought, but some of the great discoveries in the past have been made by physicians working much like a modern GP. Adhesion to ethical rules will always remain a must, and a basic knowledge of philosophy and history of science will facilitate good practice. Clinical practice incites broader reading, it gives ideas related to clinical medicine, it is a fascinating way to work with people and occasionally you may think you are of real help.
|
Declaration |
|---|
Funding: None.
Ethical approval: Not relevant.
Conflicts of interest: None.
|
Notes |
|---|
Holtedahl K. Research and practice combined—ideas for a life in general practice. Family Practice 2008; 25: 132–136.
|
References |
|---|
1 Howie J. Refining questions and hypothesis. In: Primary Care Research. Traditional and Innovative Approaches—Norton P, Stewart M, Tudiver F, Bass M, Dunn E, eds. (1991) Newbury Park, CA: Sage Publications. 13–25.
2 Grey DJP. The role of the general practitioner in the early detection of malignant disease. Trans Hunter Soc (1966) 25:121–175.
3 Holtedahl KA. [Diagnosis of cancer in general practice. I. Can diagnosis be made earlier?]. Tidsskr Nor Laegeforen (1980) 100:1219–1223.[Medline]
4 Feinstein AR, Pritchett JA, Schimpff CR. The epidemiology of cancer therapy. III. The management of imperfect data. Ann Intern Med (1969) 123:448–461.[CrossRef]
5 Holtedahl KA. [Diagnosis of cancer in general practice. II. What can the general practitioner improve? Case reports as a teaching aid]. Tidsskr Nor Laegeforen (1980) 100:1224–1226.[Medline]
6 Sackett DL HRGGTP. Clinical Epidemiology. A Basic Science for Clinical Medicine. Second (1991) Boston, MA: Little, Brown and Company.
7 Holtedahl KA. A method of calculating diagnostic indexes for possible cancer symptoms in general practice. Allgemeinmed (1990) 19:74–79.
8 Holtedahl KA. Probability revision in general practice: the cases of occult blood in stool in patients with indigestion, and daily smoking in patients who cough. Allgemeinmed (1990) 19:35–38.
9 Holtedahl K. Diagnosis of Cancer in General Practice. A Study of Delay Problems and Warning Signals of Cancer, with Implications for Public Cancer Information and for Cancer Diagnostic Strategies in General Practice. [ISM skriftserie nr. 16; 1991].
10 Holtedahl KA. Inter-observer variation on registration of signals of cancer. Scand J Prim Health Care (1987) 5:133–139.[CrossRef][Medline]
11 Holtedahl KA. The value of warning signals of cancer in general practice. Scand J Prim Health Care (1987) 5:140–143.[CrossRef][Medline]
12 Holtedahl KA. [Deficient weight gain in infants and small children. Causes and intervention in the maternal-child clinic]. Tidsskr Nor Laegeforen (1983) 103:2116–2122.[Medline]
13 Holtedahl KA, Haugslett B, Forstrøm B. Seven postural reflexes selected by Vojta. An aid for infant examination in mother-and-child clinics (In Norwegian). Tidsskr Nor Laegeforen (1985) 105:1225–1230.[Medline]
14 Holtedahl KA, Doumenc M, Steinert S, Roghell P. Patients with sexually transmitted disease: a well-defined HIV risk group in general practice? Fam Pract (1991) 8:42–47.
15 Holtedahl KA, Bo B, Hansen AH, et al. [Student education in a laboratory for general practice skills]. Tidsskr Nor Laegeforen (1999) 119:2854–2857.[Medline]
16 Murphy E, Spiegal N, Kinmonth A-L. "Will you help me with my research?" Gaining access to primary care settings and subjects. Br J Gen Pract (1992) 42:162–165.[Web of Science][Medline]
17 Christoffersen T, Holtedahl KA, Fors T, Ringberg U. Tonometry in the general practice setting (II): which cut-off point for referral—for which patients? Acta Ophthalmol (1993) 71:109–113.[Medline]
18 Christoffersen T, Fors T, Waage S, Holtedahl K. Glaucoma screening with oculokinetic perimetry in general practice: is its specificity acceptable? Eye (1995) 9:36–39.[Web of Science][Medline]
19 Holtedahl K, Hunskaar S. Prevalence, 1-year incidence and factors associated with urinary incontinence: a population based study of women 50-74 years of age in primary care. Maturitas (1998) 28:205–211.[CrossRef][Web of Science][Medline]
20 Holtedahl K, Verelst M, Schiefloe A. A population based, randomized, controlled trial of conservative treatment for urinary incontinence in women. Acta Obstet Gynecol Scand (1998) 77:671–677.[CrossRef][Web of Science][Medline]
21 Holtedahl KA, Verelst M, Schiefloe A, Hunskaar S. Usefulness of urodynamic examination in female urinary incontinence. Lessons from a population-based, randomized, controlled study of conservative treatment. Scand J Urol Nephrol (2000) 34:169–174.[CrossRef][Medline]
22 Holtedahl KA, Christoffersen T, Fors T, Ringberg U. Comparison between methods of tonometry: time for a change of approach (letter). Br J Ophthalmol (1993) 77:754–755.
23 Jensen UJ. Practice & progress. A Theory for the Modern Health-Care System (1987) Oxford: Blackwell Scientific Publications.
24 Mant J, McManus RHR. Applicability to primary care of national clinical guidelines on blood pressure lowering for people with stroke: cross sectional study. BMJ (2006) 332:635–637.
25 Rudebeck CE. General practice and the dialogue of clinical practice. On symptoms, symptom presentations, and bodily empathy (thesis). Scand J Prim Health Care (1992) ((suppl 1).
26 Malterud K. Women's wisdom—a source of knowledge for medicine (Editorial). Lancet (2006) 368:1139–1140.[CrossRef][Medline]
27 Nessa J. Talk as Medical Work. Discourse Analysis of Patient-Doctor Communication in General Practice (Thesis) (1999) Bergen, Norway: University of Bergen.
28 Miller W, Crabtree B. Primary care research: a multimethod typology and qualitative road map. In: Doing Qualitative Research—Crabtree B, Miller W, eds. (1992) Newbury Park, CA: Sage Publications. 3–28.
29 Zyzanski S, McWhinney I, Blake R, Crabtree B, Miller W. Qualitative research: perspectives on the future. In: Doing Qualitative Research—Crabtree B, Miller W, eds. (1992) Newbury Park, CA: Sage Publications. 231–248.
30 Kirkengen AL. Inscribed Bodies. Health Impact of Childhood Sexual Abuse (2001) Dordrecht, the Netherlands: Kluwer Academic Publishers.
31 Anvik T, Holtedahl K, Mikalsen H. "When patients have cancer, they stop seeing me"—the role of the general practitioner in early follow-up of patients with cancer—a qualitative study. BMC Fam Pract (2006) 7:19.
32 Holtedahl K, Hurum H. Cross-sectional study of morbidity, morbidity-associated factors and cost of treatment in Ngaoundere, Cameroon, with implications for health policy in developing countries and development assistance policy. BMC Int Health Hum Rights (2002) 2. http://www.biomedcentral.com/1472-698X/2/2.
33 Holtedahl K, Bonono L, Salpou D. The value of population campaigns offering free-of-charge HIV-testing: observational study in a town in Cameroon. Nor Epidemiol (2005) 15:159–164.
34 Holtedahl K, Salpou D, Bonono L. Lessons from HIV counselling and testing campaigns. Lancet (2007) 369:1166.[Medline]
35 Lakhani M, Baker M. Good general practitioners will continue to be essential. BMJ (2006) 332:41–43.
36 Hames C. In the eyes of the beholder: a thirty-year odyssey of research in primary care. In: Tools for Primary Care Research—Stewart M, Tudiver F, Bass M, Dunn E, Norton P, eds. (1992) Newbury Park, CA: Sage Publications. 1–13.
37 The Lancet. Is primary care a lost cause? Lancet (2003) 361:977.[CrossRef][Web of Science][Medline]
38 Mant D, Del Mar C, Glasziou P, Knottnerus A, Wallace P, van Weel C. The state of primary care research. Lancet (2004) 364:1004–1006.[CrossRef][Web of Science][Medline]
39 De Maeseneer JM, De Sutter A. Why research in family medicine? A superfluous question (letter). Ann Fam Med (2006) 2:17–22.[Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||