The work by the developing primary care team in China: a survey in two cities

Family Practice Vol. 17, No. 1 Oxford University Press 2000 Printed in Great Britain The work by the developing primary care team in China: a survey in two cities YT Wun, XQ Lu a, WN Liang a and JA Dickinson b Wun YT, Lu XQ, Liang WN and Dickinson JA. The work by the developing primary care team in China: a survey in two cities. Family Practice 2000; 17: 10 15. Background. China is in the process of converting its existing primary care resources into general practice. The infrastructure is different from that of many other countries. Objectives. We surveyed patients reasons for encounter (RFE) and the health providers diagnoses in the general practice clinics of two large northern cities in order to assess the nature of the work of these practices. Method. Practices whose staff had a short course of training in the theory and practice of the International Classification of Primary Care (ICPC) were recruited to document the RFE and diagnoses of patient encounters in two separate winter weeks. Results. The practices dealt mainly with chronic illness in older patients. Hypertension-related problems were the most frequent diagnoses, followed by upper respiratory tract infection. Patients also consulted very frequently for dizziness. Overall, there was good agreement between RFE and diagnosis in some organ systems. Conclusion. In their present form, the Chinese practices surveyed were delivering the full range of general practice care to a self-selected age group of patients. The ICPC was very useful for monitoring the work of general practice from the perspective of both the patients and the providers. Keywords. China, general practice care, ICPC, morbidity. Introduction China introduced general practice into Beijing, the capital city, in 1986, and into Tianjin, a neighbouring large industrial city, in 1992. The original healthcare system was not designed on the model of today s general practice in Western countries. Therefore, China is converting resources in the existing system into general practice and is training the medical teams for further development. For a few years, large cities in China have been running small sized primary care clinics in communes for primary medical care and disease surveillance. These clinics are often sited among residential units in Received 19 April 1999; Revised 10 August 1999; Accepted 6 September 1999. Research Committee, The Hong Kong College of Family Physicians, Hong Kong, a Training Centre of General Practice, The Capital University of Medical Sciences, Beijing and b Department of Community and Family Medicine, The Chinese University of Hong Kong, Hong Kong, China. Correspondence to YT Wun, Department of Community and Family Medicine, The Chinese University of Hong Kong, Hong Kong, China. communes with a moderately sized stable population. Each clinic is staffed by a team of five to six health workers headed by a doctor from the nearby regional hospital. Most regional hospitals also run out-patient clinics that are aimed largely at curative care of a more mobile population. Some hospitals reserve a number of beds for patients admitted through commune clinics. China has a primary care team slightly different from those of other countries. The team consists of nurses and different grades of doctors. Some doctors are graduates from university medical schools while some are high school graduates with 3 years of health/medical education and some receive 5 years of health/medical education after 3 years of high school. A small number of doctors have training in both Western and traditional Chinese medicine. The Training Centre of General Practice in Beijing has been running in-service training courses for members of the primary care teams in China since 1991. There are plans to introduce full-time post-graduate vocational training and professional examinations in the near future. Patients in large cities may choose different types of medical care under various types of medical insurance. The insurance may be paid by the individuals (private funding), 10

General practice in two Chinese cities 11 or sponsored by the government or employers (public funding). The publicly funded insurance covers out-patient and hospital care. Those patients who are covered for hospital care could and often do go directly to the hospital specialist clinics instead of to the primary care team. However, the situation has changed gradually in the past 2 3 years. The number of patients attending the communebased clinics is increasing by one-third every year. 1 It was thought to be of interest to examine the work of general practices in China at this stage of early development. The work in any practice is the result of interaction among patient characteristics, health provider characteristics and the organization of healthcare. The knowledge of what problems patients bring to the practice and what problems the primary care team identifies from the patients provides information for future planning and research. General practice in China, with its special infrastructure, may encounter a different work pattern as compared with other countries. The information is important for healthcare planning, training in general practice and patient education. Morbidity surveys reflect the nature of the work in the practices. Most surveys record the providers diagnoses and are doctor-centred in assessing the workload of the practice. However, patients reasons for encounter (RFE) constitute a practical source of patient information... useful for research and education. 2 They reflect the expectations of and demands on the practice and may be very different from doctors diagnoses. 3 It is valuable to know patients RFE in order to understand the work by the general practice team from the patient s perspective. The International Classification of Primary Care (ICPC) originates from the necessity to study and record patients expectations and the trigger for them seeking medical consultation (i.e. RFE). 2 Of the seven components of the ICPC, the first contains symptoms and complaints while the seventh component is the diagnoses specific to each organ system. ICPC names each organ system as a Chapter, e.g. chapter B for the haematological (blood) system, chapter D for the digestive system. Diseases or symptoms can be classified under organ systems. Thus, ICPC is the best tool for studying the relationship between the health seekers expectation and the health providers decision (diagnosis). The objective of this study was to survey the work in general practice in the two northern cities that were among the first to start general practice in China. In order to do so, we (i) describe the pattern of RFE of patients; (ii) describe the morbidity pattern as reported by the health providers; and (iii) analyse how these two differ from one another. Methods In 1997, ICPC2 was about to be published. Our team was involved in the Chinese translation of the new version. The translated rubric of diagnoses was validated by back translation, and was used for coding RFE and diagnoses in this survey. During a 6-week full-time course in general practice conducted in Beijing for doctors from various cities, the concept and application of RFE, coding and ICPC were introduced. The doctors from Beijing and Tianjan were invited to participate in the survey if they were the principal doctors from commune clinics in residential areas. They then received further briefing sessions and clarification on the inclusion and exclusion criteria for some codes. The aim of these sessions was to standardize the doctors coding behaviour and to stress the patient-oriented approach of the patient s RFE. The survey was done in the winter season (December 1997 March 1998). One week in early winter and one week in late winter were selected by individual practices in which to conduct the survey. Data sheets were sent to participating doctors who were to complete them after seeing each patient. The data consisted of three categories: patient s personal data; patient s RFE; and doctor s diagnosis and investigation (if any). The doctor asked for and ascertained the patient s RFE during the initial stage of each consultation. The RFE was the patient s explicit reason for seeking the consultation, as far as possible in the patient s own words, and not the doctor s interpretation of or conclusion about the reason. The RFE and the diagnosed problems of all patients seen during each of these 2 weeks were coded. If a patient returned for the same problem within the same week, that consultation was excluded. Up to three RFE and four diagnoses were allowed for each patient, on the assumption that health providers may draw more conclusions, e.g. from preventive care, apart from the patients presenting complaints, and that Chinese patients present fewer complaints than their Western counterparts due to cultural differences. One of the authors (L.X.Q.) randomly checked ~10% of the data sheets for the accuracy of coding against the RFE and diagnoses as written on the data sheets. Within the study period, frequent contacts with the participating doctors were maintained to explain any difficulties with ICPC which was new to the Chinese profession. Statistics Basic data were analysed with descriptive statistics with chi-squared and t-tests for the determination of significant differences between the two groups. Multiple regression was used for association among different variables. The statistically significant level was taken to be 0.05. Results The patient sample The survey recruited 28 commune clinics run by 42 primary care doctors (12 university graduates, 16 with

12 Family Practice an international journal 3 years of health education and 14 with 5 years of health education). There were 8595 patient encounters (Table 1), 57.9% from Beijing and 42.1% from Tianjin. They were relatively old, with a mean age (± SD) of 50.39 ± 22.3 years. Most of them (60.3%) received education beyond primary school, including 10.0% with tertiary education, while 16.7% had no formal schooling. They were mainly factory workers (43.0%) or administrators (19.2%). The vast majority (74.2%) were married. The medical expenditure was covered by public insurance in 64.5% of cases, by private funding in 34.0%, and by other means in the remainder. Only 8.7% were current smokers (at least one cigarette a day for the previous year), and 10.1% drank alcohol. Reasons for encounter and the diagnoses There were 13 454 RFE (1.6 per encounter) and 9682 diagnoses (1.1 per encounter). One reason for encounter was identified in 56.3% of the patients, while one diagnosis was made in 89.4% (Table 2). Only 5.0% of patients were referred to other specialists. When the RFE and the diagnoses are listed in the order of frequency of the organ systems, the sequence is found to be similar for both (Table 3), except for the ICPC chapter relating to the nervous system which TABLE 1 Age and sex distribution of the patient encounters Age group in years Male Female Subtotal (%) 0 9 370 349 719 (8.4) 10 19 210 176 386 (4.5) 20 29 212 258 470 (5.5) 30 39 347 421 768 (8.9) 40 49 532 722 1254 (14.6) 50 59 442 685 1127 (13.1) 60 69 911 1122 2033 (23.7) 70 79 770 712 1482 (17.2) 80 210 146 356 (4.1) Total 4004 (46.6%) 4591 (53.4%) 8595 (100%) TABLE 2 Numbers of RFE and diagnoses for each encounter No. RFE Diagnoses a 1 4842 7682 2 2647 743 3 1106 126 4 34 a There was no diagnosis in 10 encounters. TABLE 3 System codes occurring in RFE and diagnoses in descending order of frequency Order RFE (%) n = 13 454 Diagnosis (%) n = 9682 1 R respiratory (44.0) K cardiovascular (37.2) 2 K cardiovascular (25.1) R respiratory (34.6) 3 N neurological (24.0) D digestive (10.7) 4 A general (19.6) L locomotive (5.4) 5 D digestive (15.1) T endocrine (5.4) 6 L locomotive (6.9) A general (5.0) 7 T endocrine (4.7) S skin (3.0) 8 S skin (4.6) P psychological (2.6) 9 U urology (3.7) U urology (2.1) 10 P psychological (3.1) N neurological (1.6) 11 F eye (1.8) F eye (1.2) 12 H hearing (1.0) X female genital (0.9) 13 X female genital (0.9) H hearing (0.6) 14 B blood (0.3) Y male genital (0.4) 15 W reproductive (0.2) B blood (0.3) 16 Y male genital (0.2) W reproductive (0.3) 17 Z social (0.1) Z social (0.001) occupies the third place for RFE but the tenth place for diagnoses. The 10 most common individual ICPC codes, however, show a different picture from the system codes (Table 4). That the nervous system was the third most common system in RFE was due to N17 (dizziness or vertigo not due to auditory causes), amounting to 1331 patients (15.5%), and N01 (headache, 5.0%). However, on the whole, few neurological diseases were diagnosed. Agreement between patients RFE and doctors diagnoses The agreement may be considered both quantitatively and qualitatively. In 5083 patients (59.1%), the numbers of RFE and diagnoses were the same, while 3245 patients (37.9%) presented more RFE than the providers diagnoses. Only in 3% of patients did the providers make more diagnoses than the patients RFE. In 1713 patients (19.9%), the RFE and the diagnoses involved completely different ICPC chapters, i.e. the organ systems of the providers diagnoses were completely different from those of the patients RFE. A strict oneto-one comparison between RFE and diagnoses was not possible because up to three RFE and four diagnoses were allowed for recording purposes and the providers did not record them in matching sequence (first diagnosis for first RFE). The provider might have recorded the respective diagnosis for the patient s first RFE as the second diagnosis. However, in general, there was good

General practice in two Chinese cities 13 TABLE 4 The commonest 10 ICPC codes for RFE and diagnoses in descending order of frequency Order Codes for RFE (%) n = 13 454 Codes for diagnoses (%) n = 9682 1 R05, cough (18.9) R74, URTI (17.3) 2 N17, vertigo/dizziness (15.5) K86, hypertension, uncomplicated (8.7) 3 A03, fever (10.8) K91, cerebrovascular disease (5.4) 4 R21, throat symptoms (9.4) K85, elevated blood pressure (5.0) 5 N01, headache (5.0) A98, health maintenance/prevention (4.0) 6 K04, palpitation (4.4) T90, diabetes mellitus type 2 (3.8) 7 K02, chest tightness (4.3) R77, acute laryngitis/tracheitis (3.5) 8 K50, medication CVS (4.3) R79, acute bronchitis (3.2) 9 A44, preventive (3.5) K74, IHD with angina pectoris (2.9) 10 R03, wheeze (3.3) K87, hypertension, complicated (2.9) CVS = cardiovascular system; URTI = upper respiratory tract infection; IHD = ischaemic heart disease. agreement between RFE and diagnosis especially with respect to the digestive, eye, respiratory, female genital and family planning systems (all 85% agreement). There was poor agreement in three systems: general (ICPC chapter A), neurological (chapter N) and social (chapter Z). The providers made more diagnoses regarding the respiratory system in patients presenting with general complaints, the cardiovascular system for neurological symptoms, and the psychological system for social RFE. Patients with complaints of dizziness/vertigo (N17) As the RFE of N17 occurred so frequently (in 1331 or 15.5% encounters), this group of patients was analysed further. These patients were older than the whole sample, the mean age being 60.8 ± 13.5 years, and significantly different from those without N17 (mean age 48.5 ± 23.1 years, t-test P 0.001). They were also significantly different from those without N17 in terms of education (chi-square P = 0.03), marital status (chi-square P 0.001), occupation (chi-square P 0.001) and method of payment (chi-square P 0.001) but not in terms of gender (chi-square P = 0.95). Those with N17 were more likely to be older, married, factory workers, with less schooling and covered by public insurance. Logistic regression showed a significant association only with marital status, occupation and age (at P 0.001). Among these patients, N17 was the only RFE in 634 (47.6%), but the health provider made only one diagnosis in 1030 (77.4%). The most frequent diagnoses for these patients were: hypertension (K86, 420 or 31.6%), elevated blood pressure (K85, 270 or 20.3%), cerebrovascular disease (K91, 167 or 12.5%), ischaemic heart disease (K74 or K76, 102 or 7.7%) and syndrome of the cervical spine (L83, 55 or 4.1%). Thus, the majority of these patients were found to have an abnormality in blood pressure. Among those with K85, K86 and K91, 83.0% of the encounters were follow-up consultations. Discussion By placing clinics within the residential areas of communes, general practice in Beijing and Tianjin is easily accessible for patients. This accessibility particularly suits the elderly and the disabled. The present study is a survey of the problems brought by patients and identified by primary care providers in these commune practices with their special infrastructure. In particular, we identified who sought primary care there, why they consulted the team and what problems they were found to have. The findings may not reflect epidemiological data, but rather the patients response and illness behaviour for a particular primary care system. This survey shows that a very high proportion of the patients attending the commune practices are elderly (45.0% aged 60 years or over). Dizziness, respiratory symptoms, cardiovascular symptoms and requests for preventive care were the four main groups of RFE (Table 4), while hypertension-related problems (K85, K86, K87 and K91) were the most frequent morbidities. The high attendance by elderly patients results in a high frequency of chronic illness which is managed most appropriately in general practice. 4 Though upper respiratory tract infection was the most frequent disease, it accounted for only 17.3% of cases, and the bulk of the workload for the health providers was due to chronic illness, especially cardiovascular problems. The next plan of development for China should be the practice management of chronic illness and patient education on the prevention of chronic respiratory and cardiovascular diseases.

14 Family Practice an international journal It is not surprising to see that the RFE differ from the health provider s diagnoses in 20% of the encounters, as substantial mismatches have been shown by Veitch between patients RFE and health providers diagnoses. 3 In Veitch s report, the mean RFE was 1.4 per encounter, the mean diagnosis was 1.3, and both patients and providers reported the same number of ICPC rubrics in 62% of cases. Our corresponding findings are 1.6, 1.1 and 59.1%. Contrary to our initial assumption, the patients had more RFE than the diagnoses made by the health providers. The difference between the average number of RFE and that of diagnoses is relatively large. This larger difference appears to be the result of more RFE presented by patients together with less diagnoses reported by health providers. Further studies are required to confirm this hypothesis and analyse its implications. The agreement between RFE and diagnoses was very high in some organ systems, e.g. the digestive system, the eye and the respiratory system. This supports the observation 5 that, in selected medical conditions in the elderly, there is substantial agreement between self-report and the provider s report. The prominent exceptions are patients complaints relating to dizziness, the male reproductive system and social problems. Out of the 13 encounters with social complaints, the provider agreed on five and made psychological diagnoses on six. Though the number of these RFE is too small for statistical analysis, it appears that there is a large discrepancy between the patients and the providers ideas on psychosocial problems. This discrepancy may result in frustrating consultations 6 and unresolved problems. On the other hand, out of 14 RFE relating to the male reproductive system, the providers agreed on 13 but identified 20 similar diagnoses from patients who presented with other RFE. Male patients probably did not wish to consult the GPs about their reproductive problems, at least explicitly. It would be worthwhile to know whether they regarded other specialists to be better at handling these problems. The interesting observation in this survey is the predominant RFE of dizziness which was the second most frequent RFE and made the nervous system the third most involved organ system. These patients were more likely to be older in age, married, and to have received less schooling. The vast majority of them were recorded as having cardiovascular problems especially hypertension (which is mainly an asymptomatic condition) and, to a much lesser extent, ischaemic heart disease. If they were already known to have hypertension, it would be of interest to find out why they linked dizziness with elevated blood pressure. They perhaps thought that dizziness was a symptom of hypertension or even stroke. There is, of course, the possibility that these patients used dizziness as an entry ticket to have their blood pressure checked or their hypertension monitored. If this idea of an entry ticket was in fact true, the health providers met the patients overt expectation by making the appropriate diagnoses and acting accordingly. However, there is a need to organize relevant management guidelines for hypertension, stroke and ischaemic heart disease. Patient education on these diseases and their management guidelines will alleviate unnecessary anxiety, win patients co-operation in management, and promote better utilization of the healthcare system. The high rate of hypertension and cardiovascular diseases among the patients was striking. It is also striking to note the low prevalence of smokers in the study sample (8.7%) in comparison with the general population in the region (about two-thirds of males aged 25 years are smokers 7 ). It might be postulated that the discovery of a cardio- or cerebro-vasuclar disease enforced their motivation or even enabled them to quit smoking. These patients may be a suitable target group for retrospective studies of the success of quitting smoking in a country with a high prevalence of smokers. On the other hand, the heavy load on general practices due to these conditions could be lessened by promoting a cigarettefree lifestyle in the residents. Conclusion This survey describes patients problems and health providers work in the general practices developed so far in China. These practices are quite different from those in other countries. The findings still serve the same purposes as other morbidity surveys in describing the patient population and the demands on the healthcare system. The survey also reveals areas for further development of services and research for better quality and utilization of care. The commune-based general practice in these two Chinese cities served residents from the older age group and dealt mainly with chronic illness, as should be the case for general practice. The health providers satisfied most of the patients RFE. Though new to the ICPC, they found little difficulty in using the coding system and understanding the implication of RFE. The infrastructure of these practices may be adapted or modified by underdeveloped countries during their early phases in the development of general practice. ICPC could be used to monitor patients expectation from the healthcare system and their morbidity pattern. As chronic illness is more common in those of older age who are also more likely to have co-morbid problems, management guidelines and patient education on geriatric chronic illness would be very helpful for general practice in China. The younger work force may have different expectations from primary care. Their expectations should also be studied so as to extend the service of commune-based general practice to them. Of course, general practice itself may need adjustment in order to meet these expectations. References 1 Tianjin Health Bereau. Tianjin Shequ Weisheng (Tianjin Community Health). Tianjin: 1998.

General practice in two Chinese cities 15 2 WONCA Classification Committee. International Classification of Primary Care. 2nd edn. Oxford: Oxford University Press, 1998. 3 Veitch P. A comparison of patient-reported reasons for encounter and provider-reported diagnoses. Fam Pract 1995; 12: 408 412. 4 Hasler J. The very stuff of general practice. J R Coll Gen Practitioners 1985; 35: 121 127. 5 Bush T, Miller S, Golden A, Hale W. Self-report and medical record report agreement of selected medical conditions in the elderly. Am J Public Health 1989; 79: 1554 1556. 6 Arborelius E, Bremberg S, Timpka T. What is going on when the general practitioner doesn t grasp the situation? Fam Pract 1991; 8: 3 9. 7 Liu B, Peto R, Chen Z et al. Emerging tobacco hazards in China: 1. Retrospective proportional mortality study of one million deaths. Br Med J 1998; 317: 1411 1422.