The patient as a source to improve the medical record

Transcription

1 Family Practice Vol. 19, No. 2 Oxford University Press 2002 Printed in Great Britain The patient as a source to improve the medical record M Lauteslager, HJ Brouwer, J Mohrs, PJE Bindels and HGLM Grundmeijer Lauteslager M, Brouwer HJ, Mohrs J, Bindels PJE and Grundmeijer HGLM. The patient as a source to improve the medical record. Family Practice 2002; Background. The problem list is an important tool in general practice for care as well as research purposes. As the central part of the problem-oriented medical record, it lists the main medical problems which the GP wants to have knowledge of during any patient encounter. The assessment of its quality is usually made by comparing with other sources of information on the patient s problems. Objective. This study addresses the question of to what extent the problem list can be improved by asking the patient about their own medical problems. Methods. During 7 weeks, all patients who visited three GPs in a health care centre in an Amsterdam suburb were interviewed. During the interview, they were confronted with the problem list made by their own GP and stimulated to make suggestions for addition or removal of problems. Results. All in all, patients were in agreement with 88% of all listed problems. The completeness of the problem list could be increased by 28%, while 4% ultimately were removed a net gain of 24%. Conclusion. The patient can be used as a sourcetool for improvement of the quality of the problem list when its prime function is patient care. It becomes more complicated when the problem list also serves a research purpose. Clear inclusion rules will then have to be formulated. Keywords. Data quality, medical records, problem oriented. Introduction The problem list is the core of the problem-oriented electronic medical record. It provides the GP with an overview of all relevant health problems. The problem list structures the electronic record system according to the guidelines of the Problem Orientated Medical Record. 1 Originally designed for care purposes (further management of chronic problems, knowledge of what is important for the patient), the problem list also has been put to important epidemiological use, especially the electronic version. Prevalence rates can be calculated on the basis of problem list data. 2,3 For both purposes, the quality of the data in the problem list is of vital importance. Completeness and accuracy are standard indicators of the quality of electronic medical records. 4 Received 14 February 2001; Revised 14 August 2001; Accepted 1 November Department of General Practice, Academic Medical Center, Amsterdam, The Netherlands. Correspondence to HJ Brouwer, Department of General Practice, Division of Public Health, Academic Medical Centre/University of Amsterdam, Meibergdreef 15, 1105 AZ Amsterdam, The Netherlands. Incomplete and inaccurate problem lists may lead to insufficient care and preventive activities, cause communication problems with locums and hinder decision support. While incomplete problem lists ultimately cause underestimation of prevalence, inaccuracy may also cause overestimation. The assessment of completeness and accuracy implies comparing the record information with other, preferably independent, sources of information. Sources used for comparison have been paper records, 5 prescriptions, 6 dispensed medication, 7 direct observation, 8 and specialized registries. 9 The patient has rarely been used as an alternative or supplementary source of information on general practice data, 10 although the accuracy of patients self-reports on the existence of chronic diseases has been found to be adequate when compared with external assessment. 11 The present study was conducted to explore its use for improvement of medical record information. In this study, we address two questions (i) Can the patient be used as an information source to improve the completeness and accuracy of the problem list defined by the GP? 167

2 168 Family Practice an international journal (ii) When doctor and patient disagree, which reasons do they give? Methods The study took place in a health centre in Diemen-Zuid, a suburb of Amsterdam in The Netherlands. Three of the five GPs in the centre participated in the study; they provide primary care for ~3600 patients. Before the study, all computerized problem lists had been updated by the GP who manually checked these lists by comparing them with all lifetime medical correspondence (including lab results, X-ray results and specialist reports) and the medication lists of all patients. Included in the present study were all patients older than 18 years who consulted one of the three GPs during seven consecutive weeks in March and April After their consultation, in which the GP asked for and received informed consent, the patients were interviewed by the researcher in a separate room within the health centre. The researcher explained to the patient which problems are eligible for inclusion on the problem list and asked the patient the following questions (i) Do you have any chronic complaints, problems or diseases? (ii) Have you had any operations? (iii) Are you allergic to anything? (iv) Do you have an intolerance for any drugs? The researcher compared these problems on the spot with the updated problem list provided by the GP. Any discrepancies were presented immediately to the patient. The GP was confronted with discrepancies whenever time was available. In cases of persisting disagreement, the arguments were documented on a form. However, the GP made the final decision to add problems to or remove them from the list. Improvement in completeness was measured by calculating the percentage of problems which were not on the list before but were mentioned by patients and, according to the GP, should have been on that list. Improvement in accuracy was measured by calculating the percentage of those problems on the updated list which both patient and GP agreed should not have been on that list. Categorical data were summarized as proportions, and the relationships across groups were examined using contingency tables and the chi-square statistic with Bonferroni correction for multiple comparisons. The analyses were conducted using SPSS (version 8). Results Inclusion In the 7-week period, the three GPs were consulted by 441 patients. Four patients refused to participate in the study, which left 437 patients to be interviewed, 67% of whom were women. The mean age was 45 years with a range from 18 to 90. GP and patient agreement Of all 910 problems defined for these patients (Fig. 1), 55% were mentioned by them spontaneously. When presented to them by the researcher, patients agreed with another 33% (n = 304); the age distribution of these omitting patients was not different from that of all interviewed patients. Not mentioned spontaneously were problems which do not present symptoms, such as hypertension, or problems not disabling enough at the time of the study. Overall, patients agreed with 88% of the problems defined by their GP. Patient disagreement with GP Patients disagreed with 12% (n = 104) of all problems listed by their GP. These concerned psychological FIGURE 1 Problems on the GPs lists

3 The patient as a source to improve the medical record 169 problems (23%) especially anxiety disorder and depression, irritable bowel disorder (13%), musculoskeletal problems (12%), notably lower back pain, circulatory conditions (12%), especially hypertension, and various other conditions (40%). The reason patients most often (56%) gave for disagreeing was that the problem had not given them any trouble for a number of years. They also disagreed because they did not consider it a problem (17%) or even denied ever having had the problem (16%). Mainly hypertensive patients considered their problem cured since, using medication, their blood pressure was normal (10%). All of these concerned diagnoses which were recorded while they were in hospital. The GPs accepted the patient s argument in 4% of all problems (n = 36) and, considering them not relevant, removed them from the lists. In 7% of the listed problems (n = 68), patient and GP could not agree on whether the problem should be on the GP s list. These problems were not removed from the lists by the GP, usually because of the risk of recurrence (depression). Problems generated by patients A total of 260 patients mentioned 411 problems which had not been present on the GPs problem lists (Fig. 1). The GPs agreed that 255 of these (62%) should be added to the problem lists. These doctor omissions occurred significantly more frequently (P 0.05) in the age group. Most frequently added were events in the past such as appendectomy (11%), medication intolerance (10%), hysterectomy (4%), cholecystectomy (3%), allergic rhinitis (9%). The remaining 156 problems mentioned by patients were not considered chronic or severe enough (headache, tiredness) by their GP to merit addition to the problem list. Adding another 255 problems mentioned only by patients improved the completeness by 255/910 = 28%. Only 4% (n = 36) of the listed problems were removed after consulting the patients, thus further improving the accuracy of the listed problems. Thus, the net gain of asking the patient was an increase in listed problems of 24%. Clusters of problems Table 1 lists the main clusters of problems on the final problem lists after processing the opinions of patients and doctors. Seven of the 21 across-group comparisons (type of medical problem) were statistically significant. A significantly different pattern of agreement emerged comparing psychological problems with, respectively, digestive (chi-squared , P 0.01), musculoskeletal (chi-squared , P 0.01), respiratory (chi-squared , P 0.01), female genital (chi-squared , P 0.01) and other problems (chi-squared , P 0.01). For nearly half the psychological problems, the patients only agreed after being shown the GP s list. At the same time, when patients disagreed, psychological problems were kept on the list by their GPs more often in comparison with other problems. Psychological and circulatory problems were least likely to be presented only by patients. The pattern of agreement regarding circulatory problems was significantly different from digestive (chi-squared , P 0.01) and respiratory problems (chi-squared , P 0.01). Patient and doctor initially agreed on the existence of circulatory problems more often than on any other cluster of problems. One-third of the digestive (notably appendicitis) and respiratory (notably hay fever) problems were added on the initiative of the patient. Discussion This study suggests that asking the patient can be an important tool for additionally improving the quality of the problem list in general practice. In addition to sources such as medical correspondence and medication history, the patient seems to be a substantial source of TABLE 1 Origin of problems by cluster (percentages per cluster) Cluster of problems Independently listed Patient agrees after being Patient does not agree after Presented by patient only. by patient and GP confronted with GP list being confronted with Added to the list by GP GP list. Kept on the list Digestive (n = 136) Circulatory (n = 126) Musculoskeletal (n = 132) Psychological (n = 112) Respiratory (n = 98) Female genital (n = 103) Other (n = 422) Total (n = 1129) Chi-squared 18 = 120.1; P for differences across all groups of problems.

4 170 Family Practice an international journal information on the problem list and, according to the judgement of their GP, a rather good one in one-third of the 104 listed problems where the patient did not agree with them, the GP followed their opinion and removed the problem from the list. In addition to this, the GPs agreed to add nearly two-thirds of the problems only mentioned by their patients. As a result, the number of problems considered important enough to be listed increased by a quarter. These results were obtained in practices with a typical middle-class suburban population and an age distribution not different from that of The Netherlands as a whole. They should not be generalized to all patient groups. It is to be expected that elderly patients forget more problems such as hypertension, while the doctor will make an effort to have complete information for this group. Patients from a lower social class will probably be even less able than middle- and higher class patients to describe and mention problems, especially those of a psychological nature. To what extent was this result influenced by patients giving socially desirable answers? The arrangement of the interview setting was such as to minimize this as much as possible the interviewer was a medical student, and the interviews were conducted in a room separate from the consulting room. Furthermore, both the extent of disagreement (11%) and the sheer number of newly mentioned problems were quite substantial, which seems to indicate that patients felt comfortable enough to state their own opinions. Different results might be obtained with the GPs as the interviewer. Only consulting patients were interviewed. This might have biased the sample towards patients with active problems. The fact that most patients interviewed (79%) had at least one problem on the list could be an indication of this bias, as the overall percentage for patients with at least one problem on the list in the health centre was 53%. This selection bias could have underestimated the number of problems which should have been removed because they have stopped being relevant. Patients who consult their GP rarely or not at all simply cannot be reached by this method. On the other hand, the effect of sanitation of the problem list is probably greatest in the group with active problems. Most problems are lifelong (surgery, allergy, chronic disease, intolerance), and under-registration is much more important than over-reporting, as this study shows. Patients disagree with doctors especially on the recording of psychological problems such as anxiety disorders, depressions and addictions. This reflects the different significance these problems have for patients and GPs. On the one hand, patients tend to see these problems as stigmatizing and do not want to be reminded once the problem no longer requires medical attention. 12 Doctors, on the other hand, require that the problem list includes psychological problems which might be effectively cured but have a risk of recurrence. While GPs might consider a problem to be of temporary influence and without consequences, patients sometimes think otherwise. This clearly seems the case with digestive problems such as old appendicitis operations and respiratory problems such as hay fever. Circulatory problems are often diagnosed after referral and might be badly communicated to patients, hence the relatively low proportion of self-reporting by patients. Before the present study, the participating GPs had not deactivated or removed any listed problems. Feedback from the patient seems essential for this and, by definition, patients stop consulting when medical problems no longer bother them. In due course, this could inflate a problem list, leading to an overestimation of morbidity. Problem list inflation is less of a handicap for care than for epidemiological use. In studies validating the patient s self-report of the presence of chronic disease, the medical record is usually taken as the criterion standard for comparison. 1,13 Knowing that the problem list is only one part of the medical record, the three GPs participating in the study, before the interviews, scrutinized their patients medication lists and correspondence with medical specialists in order to improve the quality of the problem lists. The present study then shows that the medical record itself might have ample space for further improvement. However, improvement in terms of patient care is not necessarily the same as improvement in terms of epidemiological use. Only the chronic problems which actively exert an influence on the patient s life and are therefore in need of medical attention belong to the category where the aims of care and of research (morbidity recording) both require recording on the problem list. How clinically relevant is the quantitative net gain of 24% in the problem lists obtained by using information from the patient? Patient care surely benefits from the addition by patients of several groups of problems. Of these, medication intolerances (10% of the net gain) can have serious consequences if undetected. Not knowing about a past operation (15% of the net gain) can sometimes be an embarrassment for the doctor. Chronic complaints, i.e. headaches and back pains (15% of the net gain), should be known to the doctor and any locum because of their importance for patients. Finally, the largest group (60% of the net gain) comprised problems which were omitted by the GP for unclear reasons. The problems in this group are of varying medical significance but nevertheless seen by the GP as important to record (i.e. hypertension, several malignancies, eczemas). The patient may be considered an important source of information for optimizing the quality of the medical record. References 1 Weed LL. Medical Education and Patient Care. Cleveland Case Western Reserve University Press, 1969.

5 The patient as a source to improve the medical record Metsemakers JF, Knottnerus JA, van Schendel GJ, Kocken RJ, Limonard CB. Unlocking patients records in family practice for research, medical education and quality assurance the Registration Network Family Practices. Int J Biomed Comput 1996; van den Akker M, Buntinx F, Metsemakers JMF, Roos S, Knottnerus JA. Multimorbidity in general practice prevalence, incidence, and determinants of co-occurring chronic and recurrent diseases. J Clin Epidemiol 1998; Hogan WR, Wagner MM. Accuracy of data in computer-based patient records. J Am Med Inform Assoc 1997; Pearson N, O Brien J, Thomas H, Ewings P, Gallier L, Bussey A. Collecting morbidity data in general practice the Somerset morbidity project. Br Med J 1996; Scobie S, Basnett I, McCartney P. Can general practice data be used for needs assessments and health care planning in an inner- London district? J Public Health Med 1995; Donahue JG, Weiss ST, Goetsch MA, Livingston JM, Greineder DK, Platt R. Assessment of asthma using automated and fulltext medical records. J Asthma 1997; Stange KC, Zyzanski SJ, Smith TF et al. How valid are medical records and patient questionnaires for physician profiling and health services research. Med Care 1998; Metsemakers J. Unlocking patients records in general practice for research, medical education and quality assurance the Registration Network Family Practices. Amsterdam Thesis Publishers Amsterdam, Whitelaw FG, Nevin SL, Milne RM, Taylor RJ, Taylor MW, Watt AH. Completeness and accuracy of morbidity and repeat prescribing records held on general practice computers in Scotland. Br J Gen Pract 1996; Kriegsman DMW, Penninx BWJH, Eijk JThM van, Boeke AJP, Deeg DJH. Self-reports and general practitioner information on the presence of chronic diseases in community-dwelling elderly a study on the accuracy of patients self-reports and on determinants of inaccuracy. J Clin Epidemiol 1996; Ridsdale L, Hudd S. What do patients want and not want to see about themselves on the computer screen a qualitative study. Scand J Prim Health Care 1997; Fowles JB, Fowler EJ, Craft C. Validation of claims diagnoses and self-reported conditions compared with medical records for selected chronic diseases. J Ambulatory Care Manage 1998;