Summary

Objective:(1) To investigate the morbidity pattern of patients attending 4 government general practice (GP) clinics, (2) to generate information for strategic planning of health care services to assist decision on priority for development of quality improvement, and (3) to understand occurrence of common diseases in general practice.

Design: A cross-sectional survey from 1^st September 2004 to 31^st August 2005 of health problems/diagnoses of clinical encounters of all subjects attending 4 government GP clinics during the one year study period, coded using the International Classification of Primary Care (Revised Edition)(ICPC-2) by attending family physicians. The diagnoses or problems identified by the attending family physicians in each patient encounter were recorded. Training and clinical audits on the coding practice were organized to ensure the validity of the codes. Microsoft Visual FoxPro based programmes were used in the analysis of the data retrospectively.

Subjects: All subjects attended 4 government GP clinics during the one year study period. The four GP clinics were all Family Medicine training centres. Three of the clinics served patients who were government servants,pensioners and their dependents. The remaining one clinic served the general public of Hong Kong by referral.

Main outcome measures: The diagnoses or problems identified by the attending family physicians in each patient encounter.

Results: In the first year of implementation of ICPC-2 in 4 GP clinics, 53 314 individual patients (24 953 male and 28 361 female patients) were identified among the 166 957 clinical encounters. Each patient had an average of 1.7 problems per encounter. On average, each individual patient used the service 3.13 times per year. There was a bimodal age distribution of the patient population, with one peak at age 40-99 and another peak at age 10-14.

The top ten diagnoses in the four government GP clinics, in descending order, were upper respiratory tract infection (16.2%), uncomplicated hypertension (12.9%), non-insulin dependent diabetes mellitus (4.8%), lipid disorder (4.1%), atopic eczema (2.6%), allergic rhinitis (2.5%), dermatophytosis (1.7%), elevated blood pressure (1.7%), contact dermatitis (1.7%) and obesity (1.5%). If important cardiovascular risk factors including hypertension, diabetes, lipid disorders, smoking abuse, overweight and obesity were combined, these factors were the most common health problem among the four GP clinics and outnumbered the burden of URTI. The frequency was 72 260, which was 25.47% of all the health problems treated in the clinics, (or 43.28 per 100 patient encounters).

Conclusion: ICPC-2 is useful for our understanding of patient characteristics and morbidity pattern in the clinics. The data leads to an epidemiological understanding of the patients that utilizes the services of the four clinics. Cardiovascular risk factors are the most common health problems among the patients. This has the implication for the need to develop evidence based guidelines and clinical audit, and re-allocation of health care resources for better management of these risk factors to prevent future cardiovascular diseases. Strategic planning in health promotion of healthy eating, exercise and prevention of cardiovascular diseases may also be the priority area in health care planning.

Keywords: International Classification of Primary Care-2, morbidity pattern, cardiovascular risk factors, health care planning, clinical audit

摘要

目的：(1)調查4家公立全科醫學診所病人的發病模式；(2)為衛生服務的策略規劃提供資訊，協助確定質量改善工作的重點；和(3)瞭解全科醫療中常見疾病的發病情況。

設計：2004年9月1日至2005年8月31日為期一年的橫斷面調查，由主診家庭醫生根據《基層醫療國際分類(ICPC-2)》對研究期內到4家全科診所就診的所有病人的健康問題/診斷進行編碼。主診家庭醫生記錄每次接診所做的診斷或確定的問題。為確保編碼的有效度。還組織了有關編碼操作的培訓和臨床審核，使用以微軟Visual FoxPro為基礎的程式對資料進行回顧性分析。

對象：一年研究期間到4家全科診所就診的所有病人。4家診所均為家庭醫學培訓中心；其中3家為公務員、退休公務員及其家屬提供服務，1家為轉診來的香港普通病人提供服務。

測量內容：主診家庭醫生每次接診病人時所確定的診斷或問 題。

結果：在4家診所實施ICPC-2的第一年，共計有166957個診次，53314名病人（24953名男性和28361名女性）。每個病人每次診療過程平均有1.7個問題；每個病人年均使用衛生服務3.13次。病人人群呈雙峰式分佈，一個峰出現於40-99歲年齡組，另一個峰出現於10-14年齡組。
4家公立全科診所的前10位診斷由高到低排列：上呼吸道 感染(16.2%)、單純性高血壓(12.9%)、非胰島素依賴性糖尿病(4.8%)、血脂問題(4.1%)、特應性濕疹(2.6%)、過敏性鼻炎(2.5%)、皮癬(1.7%)、血壓升高(1.7%)、接觸性皮炎(1.7%)和肥胖(1.5%)。如果將高血壓、糖尿病、血脂問題、吸煙、超重和肥胖等重要的心血管疾病危險因素合在一起，就成為4家診所中最常見的健康問題，遠超過了上呼吸道感染所造成的負擔，其發生數量為72260次，佔診所所治療全部健康問題的25.47%，或佔全部診療過程的43.28%。

結論：ICPC-2確實對瞭解診所病人的特點和發病模式很有幫助，使我們對使用4家診所服務的病人的疾病流行情況有所瞭解。心血管疾病危險因素是病人中最常見的健康問題。這意味著需要制定循證醫學的臨床指引，施行臨床審核，並重新分配醫療衛生的資料。以便更好的管理這些危險因素，從而預防心血管疾病的發生。有關健康飲食、鍛煉和心血管疾病預防的健康教育的策略計劃，也是衛生服務規劃的優先發展的重點。

主要詞彙：基層醫療國際分類-2，發病模式，心血管疾病危險因素，衛生服務規劃，臨床審核

---

Introduction

The International Classification of Primary Care (ICPC) was published in 1987 by WONCA, the World Organisation of National Colleges, Academies, and Academic Associations of General Practitioners/Family Physicians, now known more briefly as the World Organisation of Family Doctors (Wonca).¹ In 1998 Wonca published a revised version of ICPC (ICPC-2). ICPC-2 allows classification of patients' reasons for encounters (RFE), problems/diagnoses managed, interventions, and the ordering of these data in an episode of care structure.²

The application of ICPC-2 from doing routine data recording in clinic practice is useful for a better understanding of the occurrence of diseases in the patient population, and this is important for analyzing health needs and demand in a particular setting.³ It is also important for providing information on priorities of clinical guidelines and audit. However, there has been little publication on morbidity pattern in general practice in Hong Kong since the late 90s. Without the knowledge of the morbidity pattern of the general public in Hong Kong, it is difficult to build a complete picture of the health needs and the health status at the primary healthcare level.

We implemented ICPC-2 coding of all problems/diagnoses of clinical encounters in the four clinics in 2004. In a clinical audit, we identified deficiencies in the correct and complete coding of diagnosis and health problems in the clinics and common errors in coding ICPC-2 as the preliminary step to ensure the reliability of the ICPC-2 database.⁴ After establishing a more reliable database, the database of all clinic encounters during the first year of the implementation of ICPC-2 can be used to obtain useful information.

Objective

(1)To investigate the morbidity pattern of patients attending 4 government GP clinics.

(2)To generate information for strategic planning of health care services.

(3)To provide information that will assist decisions to development of future clinical guidelines and audit.

(4)To achieve better understanding of common diseases treated in the general practice clinics.

Method

The study design was a cross-sectional study of all patients who attended four government general practice (GP) clinics during the period 1^st September 2004 to 31^st August 2005. The four GP clinics were all Family Medicine training centres. Three of the clinics served patients who were government servants, pensioners and their dependents. The remaining one clinic served the general public of Hong Kong by referral.

The outcome measures included the diagnoses/health problems of all clinical encounters during the study period.

To ensure the reliability of the database and possible bias due to incomplete and incorrect coding of diagnoses/health problems, training workshops were organized to ensure the knowledge and skills in ICPC-2 coding of all doctors in the clinics. An ICPC-2 working group of four experienced family physicians with special interest in ICPC-2 coding from each clinic was also formed to standardize the coding practice and to settle any discrepancy in coding. A clinical audit was carried out to ensure the correct and complete coding of diagnoses and health problems in our clinics and common errors in coding ICPC-2.⁴ Changes to improve the coding practice were implemented.

The data was analyzed using Microsoft Visual FoxPro. A set of computer programmes were developed using Visual FoxPro language. Health problems were coded using ICPC-2 and ranked according to their frequencies of occurrence in the one year study period. The total number of individual patients of the four clinics and number of individual patients coded for particular health problems were determined programmatically by identifying unique reference number (which is either the Hong Kong identity card number or a file number for those without a Hong Kong identity card). Consultation rates were expressed as rate per 100 clinical encounters.

Results

In all clinical encounters within the one year study period, 53 314 individual patients were identified among the 166 957 clinical encounters. 283 719 health problems were coded. There were 1.7 problems per clinical encounter. Each patient used the service 3.13 times throughout the one year period. There were 24 953 male patients and 28 361 female patients with male to female ratio of 0.88:1. The median age of the whole patient population was 45.

Compared to the general population of Hong Kong based on the mid-2005 provisional statistics of Hong Kong SAR Census and Statistical Department provisional statistics by age group and sex,⁵ there was a higher proportion of patients in the age group of 45-64 while there was a lower proportion of patients in the other age groups. As three of the clinics served government servants, pensioners and their dependents, there were differences in our patients compared to the general population. The male to female ratio of mid-2005 Hong Kong population was 0.92:1 while the ratio in our patients was 0.88:1. Chart 1 compares the population of Hong Kong of mid-2005 to the patients of the four clinics.

Table 1: Overall morbidity pattern by rate per 100 patient encounters and the percentage of total health problems in the 17 chapters of ICPC-2

Rate per 100 patient encounters   Percentage of total problems % Respiratory   38.9   22.9 Circulatory   30.8   18.1 Metabolic/Endocrine/Nutritional   22.5   13.2 Skin   19.4   11.4 Musculoskeletal   16.1   9.5 Digestive   11.7   6.9 Psychological   5.4   3.2 Eye   5.4   3.2 General   3.9   2.3 Female   3.5   2.1 Neurological   3.0   1.7 Urinary   2.4   1.4 Ear   2.2   1.3 Social   2.1   1.2 Male   1.8   1.1 Blood   0.8   0.5 Pregnancy   0.2   0.1

Unlike the morbidity pattern in the 1994 survey conducted by Hong Kong College of Family Physicians (HKCFP),⁶ which revealed a higher proportion of respiratory and digestive problems, our survey shows a much higher proportion of circulatory problems and endocrine problems. However, since the settings of the two surveys were different, direct comparison is not suitable and may be misleading.

Table 2: Morbidity pattern of HKCFP 1994 morbidity survey and our survey

HKCFP 1994 survey (%)   Our survey 2004-05 (%) General**   3.4     2.3   Blood   0.4     0.5   Digestive**   10.4     6.9   Eye**   2.2     3.2   Ear   1.3     1.3   Circulatory**   8.8     18.1   Musculoskeletal**   6.4     9.5   Neurological*   1.9     1.7   Psychological**   1.8     3.2   Respiratory**   47.1     22.9   Skin**   7.5     11.4   Metabolic/Endocrine/Nutritional**   4.4     13.2   Urinary**   1.2     1.4   Pregnancy**   0.9     0.1   Female**   1.4     2.1   Male**   0.6     1.1   Social**   0     1.2                 * indicates p<0.05 by Chi Square test ** indicates p<0.001 by Chi Square test

Table 3 illustrates the morbidity pattern by age group and by percentage of total health problems. The more advanced age groups had more circulatory and endocrine problems while the younger age groups had more respiratory problems. This probably reflected a higher proportion of acute episodic illnesses among younger people while older people had more chronic conditions such as hypertension, dyslipidaemia and diabetes.

Table 4illustrates the top 80% health problems among the patients. Apparently, upper respiratory tract infection (URTI) was the most common health problem. However, If all important cardiovascular risk factors for predicting absolute cardiovascular risk for primary prevention of cardiovascular events were combined, including hypertension complicated and uncomplicated, elevated blood pressure, diabetes, lipid disorder, overweight, obesity and smoking, these factors were the most common health problems among our patients and outnumbered the burden of URTI. The frequency was 72 260, which was 25.5% of all the health problems, or 43.3 per 100 patient encounters. This was also reflected in the percentage of cardiovascular diagnoses (alpha code K in ICPC-2 chapters) in our population, which accounted for 18.1% of all health problems coded.

The top 10 health problems in each age group are shown in Table 5.

Paediatric patients aged 0-9 had the highest proportion of URTI. Other problems such as eczema and allergic rhinitis, other infections such as acute bronchits/bronchilitis, gastroenteritis were also common.

Among adolescent patients aged 10-19, URTI was still the most frequent diagnosis, even though it was not as common as in the paediatric patients. Infectious diseases including URTI and gastroenteritis, allergic problems such as eczema and allergic rhinitis and acne were among the most common health problems in this age group.

Among adults aged 20-44, URTI was still the commonest diagnosis but not as common as among the younger age groups. Other conditions such as allergic rhinitis, eczema, infectious diseases such as gastroenteritis were among the commonest diagnosis. It was also observed that patients with musculoskeletal problems such as tenosynovitis/bursitis were more common among patients of this age group than among paediatric and adolescent patients. Cardiovascular risk factors such as obesity, diabetes and elevated blood pressure began to show on the list of common diagnoses.

Adults aged 45-64 formed the largest proportion of the patient load. Acute infectious problems and allergic problems were less common compared with the younger age groups. However, the most striking feature of this age group was the high percentage of cardiovascular risk factors such as hypertension, diabetes, lipid disorder and obesity. This may lead to future cardiovascular complications as the population ages.

In the elderly age group aged 65 or above, chronic and degenerative conditions like hypertension, diabetes, dyslipidaemia, gout, osteoarthrosis and benign prostatetic hypertrophy were important diagnoses.

Discussion

The purpose of this study has been to provide concrete information on which strategic planning of the public health care can be based. While this study only illustrates the morbidity of approximately 50 000 patients presented to the four GP clinics within a one year period, its findings provide significant implications in the direction of services our public clinics need to take in the near future.

The percentage of cardiovascular diagnoses (alpha code K in ICPC-2 chapters) among the patients in our survey illustrates cardiovascular risk factors, (including hypertension complicated and uncomplicated, elevated blood pressure, diabetes, lipid disorder, obesity, overweight and smoking) have exceeded URTI as the most prevalent health threats among the patients. As the population ages, it is estimated that, in ten to twenty years' time, these health threats will not only develop into a huge burden on the medical costs but they will also lead to enormous income loss in the community as a whole. Therefore, it is essential that the emphasis of our health care should be on the prevention of future cardiovascular events, such as stroke and ischaemic heart disease, and their consequences. In order to do this, it is highly recommended that resources should be re-allocated to the promotion of a healthy lifestyle to the public/patients.

Our study reveals difference between the age of the subject patients and the frequency of cardiovascular risk factors. It shows that there is a gradual increase in the cardiovascular risk factors and a decrease in acute episodic illness in the morbidity pattern as the groups advance in age. This may have two implications for our health care sector: (a) there is need in better management of cardiovascular risk factors among the younger population, especially the age-group between 40 and 44 to prevent future cardiovascular events, and (b) health care workers require a better knowledge and more advanced skills in geriatric medicine to deal with health problems of the growing geriatric population.

However, there are several limitations in our survey.

1. The age/sex distribution of the subject patients of this study may not be representative enough for the general population of Hong Kong because they were mainly government servants, pensioners and their dependents and some patients referred to a Family Medicine training centre even though we analyzed our data by age subgroups to partially compensate for the problem.
2. The morbidity pattern emerging from our study may have been affected by doctor-initiated follow-ups as it is the practice of our clinics to schedule follow-up appointments for most of our patients with chronic health problems. This may have contributed to such health conditions to appear more frequently in the study.
3. Although a clinical audit on coding of ICPC-2 was employed to ensure the accuracy of the coding process, the database used in this study was limited to the first year of our full implementation of ICPC-2. There was a possibility that the data in the initial phase of our implementation could be more prone to errors.
4. Grouping of ICPC-2 coding has its intrinsic limitation of unavailability of specific codes for some less common diagnosis such as the term Non-Insulin Dependent Diabetes, which has been replaced by Type 2 Diabetes in the medical literature nowadays.

Conclusion

ICPC-2 coding is useful in our understanding of patient characteristics and morbidity pattern. Through implementing the ICPC-2 coding system, we have noticed that cardiovascular risk factors are the most important health risks among the patients of our four GP clinics and the occurrences of such factors become more frequent in the more senior age group. However, there is still a need for another morbidity survey of a larger scale which includes both the private and public sectors so that we can gain a more thorough understanding in the morbidity pattern of Hong Kong's population which is essential for an effective allocation of our health care resources in the near future.

Key messages

1. The application of ICPC-2 coding from doing routine data recording in clinic practice is useful for a better understanding of the occurrence of diseases in the patient population, and this is important for analyzing health needs and demand in a particular setting.
2. The percentage of cardiovascular diagnoses (alpha code K in ICPC-2 chapters) among the patients in our survey illustrates cardiovascular risk factors, (including hypertension complicated and uncomplicated, elevated blood pressure, diabetes, lipid disorder, obesity, overweight and smoking) have exceeded URTI as the most prevalent health threats among the patients.
3. Our survey shows that there is a gradual increase in the cardiovascular risk factors and a decrease in acute episodic illness in the morbidity pattern as the groups advance in age.
4. There is need in better management of cardiovascular risk factors among the younger population, especially the age-group between 40 and 44 to prevent future cardiovascular events.
5. Health care workers require a better knowledge and more advanced skills in geriatric medicine to deal with health problems of the growing geriatric population.
6. There is still a need for another morbidity survey of a larger scale which includes both the private and public sectors so that we can gain a more thorough understanding in the morbidity pattern of Hong Kong's population.

---

---

References