Family Practice Vol. 17, No. 6, 514-521
© Oxford University Press 2000
Investigation and therapy in patients with different types of dyspepsia: a 3 year follow-up study from general practice
Centre of Preventive Medicine, Glostrup University Hospital, Copenhagen, Denmark.
Correspondence to V. Meineche-Schmidt, Christiansholmsvej 5, DK-2930 Klampenborg, Denmark.
Meineche-Schmidt V and Jørgensen T. Investigation and therapy in patients with different types of dyspepsia. Family Practice 2000; 17: 514–521.
Received 11 April 2000; Revised 17 July 2000; Accepted 17 July 2000.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Background. Decisions by GPs on investigation and treatment are based on the symptoms presented by the patient. The relevance of dyspepsia subgroups has been questioned, but their value in general practice has not been tested.
Objective. The aim of this study was to investigate how dividing dyspepsia into different subgroups (ulcer-like, reflux-like, dysmotility-like, uncharacteristic and relapsing dyspepsia) affected the approach of GPs to patients with dyspeptic complaints.
Methods. A random sample of GPs' patients consulting for different dyspepsia subtypes were studied by postal questionnaires 3 years after the initial consultation, obtaining information from the GPs' records on investigations, prescriptions of dyspepsia medication and gastrointestinal morbidity.
Results. In the 3 years studied, 48% of the patients were prescribed dyspepsia medication, 14% were endoscoped and 3% were referred to a specialist. The dyspepsia subtype was significantly related to the type of drug prescribed, but not to investigations or referrals. Ulcer-like and reflux-like dyspepsia were treated in the same way.
Discussion. Dyspepsia subtypes significantly influenced the treatment. Danish GPs treat all acid-related dyspepsia in the same way, and differently from other types of dyspepsia.
Keywords. Drug treatment, dyspepsia, dyspepsia subtypes, general practice, investigations.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The approach to dyspepsia in primary health care is based on symptoms presented to the doctor by the patient. The options for the GP are reassurance, watchful waiting, lifestyle advice, empirical treatment, or investigations, either in his or her own office or by referral to specialists. Although it has been suggested that such patients should be referred for endoscopy promptly1 or tested for Helicobacter pylori infection2 at first presentations, most GPs still base their approach on the reported symptoms.3–6 Certain clusters of symptoms have been defined as dyspepsia subgroups7–10 in order to guide empirical treatment. However, this approach has been disappointing when applied to patients in secondary care and in population-based studies, because of overlap of symptoms11,12 and lack of relation between subgroups and morphological findings at endoscopy.13 Muris et al.14 have reviewed the value of symptoms as guidance in the decision to refer to investigation or prescribe treatment in the GP's office. They identified certain clusters of symptoms as predictors of organic disease, but the data were not based on prospective studies. No study has been published in which the consequence of subgrouping dyspepsia, based on symptoms presented by the patient in primary health care, has been prospectively evaluated as a tool for investigation, treatment and prognosis.
The present study was a 3 year prospective follow-up of a randomly selected cohort of patients who had consulted their GPs because of dyspeptic complaints. The aim of the study was to focus on GPs' treatment and investigations over a 3 year period in relation to different dyspepsia subgroups, and to register the number of gastrointestinal diagnoses obtained during this period.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
During the period June 1991 to May 1993, a diagnostic chart was filled in, based on a structured interview by the GP, for consecutive patients consulting their GP because of dyspepsia (defined as pain or discomfort in the abdomen, judged by the GP to be related to the gastrointestinal tract). Information about 18 dyspepsia symptoms (epigastric pain, heartburn, upper abdominal pain, other abdominal pain, pain in the morning, at night and after meals, pain relieved by food, antacids, vomiting or passing of stools, acid regurgitation, nausea, morning vomiting, bloating, constipation, loose stools and incomplete evacuation) and six alarm symptoms (anaemia, jaundice, blood in stools, black stools, dysphagia and unintended weight loss) was collected.
Based on the symptoms presented, every patient was classified into a dyspepsia subgroup according to the recommendations of an international working party.8 The subgroups were: dysmotility-like, ulcer-like, reflux-like and uncharacteristic dyspepsia. Patients with more than one episode of dyspepsia were termed as having relapsing dyspepsia. The patients were investigated and treated according to the GPs' normal routines. In all, 93 GPs covering 123 610 persons aged 
18 years and covered by the National Health Security system participated and a total of 7274 diagnostic charts were obtained. This study is reported in detail elsewhere.15
In 1994, the GPs who had filled in the diagnostic charts were asked to participate in a follow-up study based on postal questionnaires. The invitation was accepted by 82 (88%) of the GPs.
Of the patients who had been observed for at least 12 months, 300 from each subgroup were randomly selected. As <300 patients were classified with relapsing and uncharacteristic dyspepsia, all patients in these subgroups were included. A few patients were registered twice (because of inconsistency of initials and change in age), so 297 charts from the dysmotility-like, 299 from the reflux-like, 296 from the ulcer-like, 112 from the uncharacteristic and 211 from the relapsing subgroups were eventually selected. A total of 170 charts were registered by doctors who declined to participate in the study and, as the identity of the patients were not known to the study group, it was not possible to trace these patients. However, the age and gender distribution and the distribution according to dyspepsia subgroups showed no significant differences between participating and non-participating doctors. The questionnaires were sent out in November 1994 and returned by April 1995. Reminders were sent out twice. The follow-up study period was the time between the date of the diagnostic chart and the date when questionnaires were filled in. The study period varied from 1.86 to 4.33 years (median: 3.11 years).
The following information was recorded: age, gender, dyspepsia subtype, alarm symptoms present, dwelling (rural, suburban or urban), the present status of the patient, number of consultations due to dyspeptic complaints, prescription of and number of weeks on medication against dyspepsia (antacids, H2 blockers, proton pump inhibitors or prokinetics) and number of referrals to specialist or hospital (to endoscopy, X-ray, ultrasound or operation) due to dyspeptic complaints. For patients who had died, the medical history and the cause of death were recorded.
The returned questionnaires were checked, and an attempt was made to get missing information by telephone contact with the GP. Investigations and prescriptions were studied in relation to the patient subtype at the initial consultation.
A total of 1045 questionnaires were sent out and 988 (95%) were returned. Patients' characteristics are listed in Table 1
.
|
Statistics
All data were transferred to Paradox 4.0. Medication was coded according to the ATC system and diagnoses according to the ICPC system. The observation period was calculated for each patient. Statistical calculations were performed with SPSS 7.0 for Windows.
Between-group comparisons were done using the chi-square test. Factors associated with the course of dyspepsia were studied using multiple logistic regression. Age was tested for linearity and age quartiles were used if linearity was not documented. Age and gender were tested for interaction and males and females were analysed separately if interaction was found. Other variables were tested adjusted for age and gender, and interaction between variables was tested. Each variable was tested in a ‘univariate’ (adjusted for age and gender) analysis. Variables with a P-value of <0.25 were then entered in a multiple logistic regression analysis with stepwise backward elimination, only leaving significant variables in the model. Ninety-five per cent confidence intervals were used.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
During the 3 year observation period, a total of 650 courses of antacids, H2 blockers, proton pump inhibitors or prokinetics were prescribed to 471 (48%) of the 988 patients. Patients with reflux-like and ulcer-like dyspepsia were prescribed medication significantly more often than patients with dysmotility-like or uncharacteristic dyspepsia. More than eight out of 10 patients with relapsing dyspepsia were prescribed medication (Table 2
).
|
H2 blockers were most often prescribed (40% of the treatments) followed by proton pump inhibitors (30%), prokinetics (16%) and antacids (14%). Prescription of antacids was unrelated to age, whereas prokinetics were prescribed significantly more often to patients aged 51–64 years and H2 blockers and proton pump inhibitors to patients aged
51 years, compared with other ages (Table 3). Women were prescribed antacids and prokinetics more often and proton pump inhibitors less often than men. No difference was found between men and women in the prescription of H2 blockers. Prescription of antacids was unrelated to dyspepsia subtype at baseline, whereas H2 blockers and proton pump inhibitors were prescribed more often and prokinetics less often in ulcer-like and reflux-like dyspepsia than in the other subtypes. Relapsing dyspepsia was significantly related to prescription of H2 blockers and proton pump inhibitors. Endoscopy was significantly related to prescription of all kinds of dyspepsia drugs (Table 3). During the observation period, 47% of prescribed medication was given for <4 weeks, 22% for 5–12 weeks, 20% for 12–52 weeks and 11% for 52 weeks (Table 4). The duration of treatment was related neither to the type of drug (Table 4) nor to the type of dyspepsia (data not shown), although there was a tendency towards shorter treatments in uncharacteristic dyspepsia and to longer treatments in relapsing dyspepsia compared with the other subgroups.
|
|
During the observation period, 30 patients (3%) were referred to a specialist (for diagnostic and therapeutic advice), 140 (14%) were endoscoped (by a specialist, but without therapeutic advice), 36 (4%) were referred to X-ray and 18 (2%) to ultrasound and 46 (5%) underwent surgery (Table 5
). Operations were significantly related to high age, whereas referrals to specialist, endoscopy or X-ray were unrelated to age. Women were significantly more often referred to specialists and less often to endoscopy compared with men. Whether the patient was from a rural, urban or suburban area was unrelated to referral pattern. Ulcer-like and reflux-like dyspepsia were inversely related to surgery, and the dyspepsia subtype at baseline was unrelated to referrals or investigations. The presence of alarm symptoms at baseline was related to surgery, but not to investigations or referrals. Prescriptions of drugs to treat dyspepsia were significantly related to referral to specialists, endoscopy and X-ray, but not to surgery (Table 6).
|
|
In 693 patients, (70%) no attempt was made to obtain a morphological diagnosis. In patients for whom a diagnosis was sought, 47 (16%) presented with normal findings, 37 (12%) with peptic ulcer and 15 (5%) with malignancy, and 132 (44%) had minor gastrointestinal findings (including gall stones and irritable bowel syndrome). In 65 patients (23%) a diagnosis not related to the gastrointestinal tract was established. Diagnoses obtained during the observation period were not related to the dyspepsia subtype at baseline (Table 7
).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The overall prescription pattern during the 3 year study period showed that nearly half of patients consulting a GP because of dyspepsia had been prescribed medication. Even though these are cumulative figures they are lower than those reported in studies from Finland3 and the UK,4 where 75% and 93%, respectively, of patients consulting their GP because of dyspepsia were prescribed drugs at the first consultation. In the Finnish study,3 42% of these prescriptions were for sucralfate, 26% for H2 blockers or proton pump inhibitors, 24% for antacids and 8% for prokinetics, whereas in the UK study,4 H2 blockers were prescribed to 36% of patients, proton pump inhibitors to 24%, antacids to 24% and prokinetics to 8%. These differences probably reflect different treatment traditions; similar findings have been described in multinational studies.16
The prescription patterns in reflux-like and ulcerlike dyspepsia in this study were very much the same, indicating that, from a therapeutic point of view, GPs treat all ‘acid’-related dyspepsia in the same manner. Dysmotility-like dyspepsia and uncharacteristic dyspepsia were treated differently and fewer patients in these categories were prescribed drugs. This might reflect differences in the expected effect of available drugs on different dyspeptic symptoms, rather than need for treatment as such. Patients with relapsing episodes of dyspepsia showed remarkable resemblance to ‘acid’-related dyspepsia, which is in accordance with the dyspepsia subtypes being predominantly ulcer-like or reflux-like in these patients.17 Proton pump inhibitors were used more often in men than in women. We have no explanation for this, but one study has shown that females have a poorer response to treatment with omeprazole than males.18 In one UK study, H2 blockers and proton pump inhibitors were prescribed more often for men than for women.19
Prescription of medication of any kind was found to be associated with endoscopy. A reducing effect on medication following endoscopy has been reported in a retrospective study from general practice in the UK20 and in a randomized trial from secondary care in Denmark.1 Our prospective study, however, could not confirm this, and nor could two Dutch studies from primary care.21,22 Although the recommended duration of treatment according to indication varies between drugs, the duration of treatment is remarkably consistent, as half of the patients were treated for <4 weeks within 3 years, one quarter were treated for 1–3 months and one quarter for 3 months within each category of drug. A Danish study from 199823 on prescriptions of H2 blockers or proton pump inhibitors in general practice in 1 year demonstrated that 36% of prescriptions were for 2 weeks only and 29% for 3 months, which is consistent with the findings in this study, even though these drugs have been used increasingly over time.24 The dyspepsia subtype seemed to have a significant effect on whether treatment was initiated and on which drug was chosen, but once treatment had started the duration was unaffected by both the diagnosis (subtype) and the drug used. Possible explanations for this finding could be that treatment of dyspeptic complaints with different drugs has given high placebo responses (
60%), and that almost half of the prescribed courses lasted 1 month, which indicates prescription of the smallest package of drug, leaving no room for differences between the drugs.
Referral to endoscopy, the commonest of all referrals, was done in 140 (14%) of the patients and was significantly related to reflux-like, ulcer-like and relapsing dyspepsia. Compared with the findings of other studies, these figures are low. Ninety-five per cent of Danish GPs5 prefer empirical treatment over prompt endoscopy on the first presentation and the vast majority of patients in this study had only one episode of dyspepsia recorded. In Finland, 14–17% of the patients were endoscoped at first presentation3 and in the UK, 74% of H2 blocker-treated patients and 63% of proton pump inhibitor-treated patients had been endoscoped.24–26 It is not surprising that endoscopy is used more often in the UK than in Denmark, as endoscopy can be performed by GPs in the UK, whereas it requires a specialist referral in Denmark. Endoscopy was performed less often in women than in men, and more women were referred to specialists, which might reflect the fact that the symptom presentation in female patients can be more difficult to interpret. Prescription of dyspepsia medication was related to referrals to specialists, endoscopy and X-ray, which reflects the fact that these options are used in patients with relapsing symptoms, patients without relevant response to therapy or patients with complicated symptoms.5 No relation was found between the dyspepsia subtype at baseline and referrals or investigations, which is in accordance with the fact that empirical treatment is the main concern of Danish GPs. The GP aims to estimate the risk of serious disease and the need for symptomatic treatment; diagnostic work-up is not their primary concern. The presence of alarm symptoms at baseline increased the likelihood of investigations or referrals. This, however, did not reach statistical significance, except in the case of referrals for surgery.
No difference was observed between patients from rural, suburban or urban dwellings. This indicates that referrals were based on the symptoms presented and were not dependent on which investigations were available.
Among the patients studied here, the incidence of serious gastrointestinal diseases was low: 1% developed a malignancy and 4% a peptic ulcer. Of those who were endoscoped, 12% had ulcers and 5% cancer. These diagnoses were obtained over a period of 3 years and therefore would be expected to be higher than findings of cross-sectional studies from open-access endoscopy units, where it has been reported that 13–16% of patients have ulcers and 1–2% cancers.27–29
In conclusion, this prospective study demonstrated that the subgroup of dyspepsia significantly influences the treatment given, but not the decision for referral, by GPs. Patients with reflux-like and ulcer-like dyspepsia were treated in the same way and differently from patients with non-acid-related dyspepsia. Development of serious gastrointestinal disease was rare.
|
Acknowledgments |
|---|
The study was supported by grants from The Public Health Insurance in Denmark (11/208-94, 11/069-97 and 11/266-98) and the General Practitioners' Research Foundation (FF-2-01-156/94-98).
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1 Bytzer P, Hansen JM, Schaffelitzky de Muckadell OB. Empirical H2-blocker therapy or prompt endoscopy in management of dyspepsia. Lancet 1994; 343: 811–816.[ISI][Medline]
2 McColl KEL, El-Nujumi A, El-Omar E and Dickson A. Evidence supporting an H. pylori test and treat startegy for simple dyspepsia. Gastroenterology [date]; 110: A28.
3 Heikkinen M, Pikkarainen P, Takala J, Julkunen R. General practitioners' approach to dyspepsia. Scand J Gastroenterol 1996; 31: 648–653.[ISI][Medline]
4 Bodger K, Daly MJ, Heatley RV. Prescribing patterns for dyspepsia in primary care: a prospective study of selected general practitioners. Aliment Pharmacol Ther 1996; 10: 889–895.[ISI][Medline]
5 Folkersen BH, Larsen B, Qvist P. General practitioners handling of patients with dyspepsia. Ugeskr Læger 1997; 159: 3777–3781.[Medline]
6 Asante MA, Patel P, Mendall M, Jazrawi R, Northfield TC. The impact of direct access endoscopy, Helicobacter pylori near patient testing and acid suppressants on the management of dyspepsia in general practice. Int J Clin Pract 1997; 51: 497–499.[ISI][Medline]
7 Kay L and Jørgensen T. Epidemiology of upper dyspepsia in a random population. Scand J Gastroenterol 1994; 29: 1–6.
8 Colin-Jones DG, Bloom B, Bodemar G et al. Management of dyspepsia: report of a working party. Lancet 1988; 337: 576–579.
9 Drossman DA, Thompson WG, Talley NJ, Funch-Jensen P, Janssens J, Whitehead WE. Identification of subgroups of functional gastrointestinal disorders. Gastroenterol Int 1990; 3: 159–172.
10 Talley NJ, Colin-Jones DG, Koch KL, Koch M, Nyren O, Stanghellini V. Functional dyspepsia: a classification with guidelines for diagnosis and management. Gastroenterol Int 1991; 4: 145–160.
11 Talley NJ, Zinsmeister AR, Schleck CD, Melton LJ. Dyspepsia and dyspepsia subgroups: a population-based study. Gastroenterology 1992; 102: 1259–1268.[ISI][Medline]
12 Agreus L, Svärdsudd K, Nyren O, Tibblin G. Irritable bowel syndrome and dyspepsia in the general population: overlap and lack of stability over time. Gastroenterology 1995; 109: 671–680.[ISI][Medline]
13 Talley NJ, Weaver AL, Tesmer DL, Zinsmeister AR. Lack of discriminant value of dyspepsia subgroups in patients referred for upper endoscopy. Gastroenterology 1993; 105: 1378.[ISI][Medline]
14 Muris JWM, Starmans R, Pop P, Crebolder HFJM, Knottnerus JA. Discriminant value of symptoms in patients with dyspepsia. Fam Pract 1994; 38: 139–143.
15 Meineche-Schmidt V, Krag E. Dyspepsia in general practice in Denmark. A 1-year analysis of consulters in general practice. Scand J Prim Health Care 1998; 16: 216–221.[ISI][Medline]
16 Meineche-Schmidt V, Talley NJ, Pap A, Kordecki H, Ohlsson L, Wahlqvist P et al. Impact of functional dyspepsia on quality of life and health care consumption after cessation of antisecretory treatment. Scand J Gastroenterol 1999; 34: 566–574.[ISI][Medline]
17 Meineche-Schmidt V, Jørgensen T. Dyspepsia subtypes in general practice—fluctuation over time. Submitted.
18 Meineche-Schmidt V, Christensen E. Which patients will benefit from omeprazole treatment? Am J Gastroenterol 2000; 95: 2777–2783.[ISI][Medline]
19 Roberts SJ, Bateman DN. Prescribing of antacids and ulcer-healing drugs in primary care in the north of England. Aliment Pharmacol Ther 1995; 9: 137–143.[ISI][Medline]
20 Hungin AP, Thomas PR, Bramble MG et al. What happens to patients following open access gastroscopy? Br J Gen Pract 1994; 44: 519–521.[ISI][Medline]
21 Lewin-van den Broek NT. Diagnostic and therapeutic strategies for dyspepsia in primary care. Thesis, Utrecht University, The Netherlands.
22 Laheij RJF, Severens JL, Lisdonk EHV, Verbeek ALM, Jansen JBMJ. Randomized controlled trial of omeprazole or endoscopy in patients with persistent dyspepsia: a cost-effectiveness analysis. Aliment Pharmacol Ther 1998; 12: 1249–1256.[ISI][Medline]
23 Meineche-Schmidt V, Fly G. Helicobacter pylori test-and-treat strategy in dyspepsia patients in general practice. Too many patients to treat? Gut 1999; 43 (Suppl.): A355.
24
Bashford JN, Norwood J, Chapman SR. Why are patients prescribed proton pump inhibitors? Retrospective analysis of link between morbidity and prescribing in the General Practice Research Database. Br Med J 1998; 317: 452–456.
25
Gilliland AE, Mills KA, Irwin WG, Steele K. Patients on H2-receptor antagonists—are we investigating them? Fam Pract 1990; 7: 43–46.
26 Goudie BM, McKenzie PE, Cipriano J, Griffin EM, Murray FE. Repeat prescribing of ulcer healing drugs in general practice—prevalence and underlying diagnosis. Aliment Pharmacol Ther 1996; 10: 147–150.[ISI][Medline]
27 Heikkinen M, Pikkarainen PH, Takala J et al. Etiology of dyspepsia: four hundred unselected consecutive patients in general practice. Scand J Gastroenterol 1995; 30: 519–523.[ISI][Medline]
28 Kagevi I, Lofstedt S, Persson LG. Endoscopic findings and diagnoses in unselected dyspeptic patients at a primary health care centre. Scand J Gastroenterol 1989; 24: 145–150.[ISI][Medline]
29 Hansen JM, Bytzer P, Schaffelitzky De Muckadell OB. Management of dyspeptic patients in primary care. Scand J Gastroenterol 1998; 33: 799–805.[ISI][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||