The previous article (Part 1) in this series1 describes the historical transition of Hong Kong’s public hospital system from being part of the Civil Service to a separate corporatised entity established by statute, charged with modernising its management and service quality. After the Hong Kong Hospital Authority (HA) was inaugurated on 1 December 1990, it took merely 12 months of preparation for the takeover of the entire public hospital system overnight on 1 December 1991. This article describes how the Authority turned the ambitious blueprint laid down in the Scott Report,2 further modified and elaborated in the Provisional Hospital Authority (PHA) Report,3 into reality.4 Current HA staff may be amazed that so many systems and processes that have long been taken for granted were once non-existent. This historical account may give not only an understanding of how the existing practices came about, but also a useful case study in healthcare organisational management.

For the first time, the public hospital system had a mission statement as laid down by the HA Board after a 3-day workshop in February 1991,5 developed according to the organisational function spelt out in the HA Ordinance (Table 1). This rather lengthy mission statement merely reiterates upfront the communitarian healthcare policy of the Government, and goes on to enunciate the Authority’s responsibility towards each of the major stakeholder parties.

From this first version of the mission statement, major strategies were derived, which served to focus the whole organisation on priority goals and targets to improve population health and service quality. This was a far cry from the past, when there were hardly any coordinated directions beyond simple bed-to-population or manpower ratios to cope with population growth, and where service improvements were heavily influenced by the preferences of powerful medical consultants and their respective political clout in securing resources within the former Medical and Health Department.

Empowered by the HA Ordinance, the Government-appointed Members of the HA (commonly referred to collectively as the HA Board) governed all public hospitals independently of the Civil Service, thus enjoying much greater flexibility in the use of available funds and the management of human resources, while accountable to the Secretary for Health and Welfare. The Chairman of the PHA, Sir Sze-yuen Chung 鍾士元爵士, was reappointed as Chairman of the HA (Table 26).

Eight functional committees under the HA Board served to introduce external expertise and steer managerial modernisation in specific areas. The chairmen and vice-chairmen of these committees together with the three principal officers constituted an Executive Committee, which was the chief decision-making body of the Authority chaired by the HA Chairman. In addition, an independent Public Complaints Committee was established to deal with appeal cases not settled at the hospitals.

At the second level, three Regional Advisory Committees were established to enable community participation. At the third level, each hospital was governed by a Hospital Governing Committee with the introduction of community leaders. For Schedule 2 (ex-subvented) hospitals, the parent body would nominate a two-thirds majority of the members (with the rest appointed by the HA) and would chair the Hospital Governing Committee.

The previous three-tiered structure of the Medical and Health Department—Headquarters, Regional Offices, and Hospitals—was simplified into two levels to enhance efficiency and hospital autonomy. The Regional Offices were abolished with their staff absorbed into the HA Head Office.

A tripartite top management structure of principal officers—Director of Operations, Chief Development Officer, and Secretary General—was initially adopted, reporting to the Executive Committee of the HA Board, until a suitable candidate for the Chief Executive (CE) position emerged.

The appointment of the three principal officers was completed upon the HA’s establishment on 1 December 1990. There was a balanced mix of talents. The Director of Operations, Dr EK Yeoh 楊永強, was a senior medical consultant recruited from within the system. The Chief Development Officer, Mr SM Pang 彭秀文, was an experienced hospital administrator recruited from Australia. The Secretary General, Mr John Chambers 湛保庶, was former Secretary for Health and Welfare from the Administrative Officer rank of the Civil Service. Recruitment of various deputies was also completed by the time the HA formally took over the operations of all hospitals on 1 December 1991. Dr Yeoh was eventually appointed to the CE position in 1994.

Under the new management structure, each hospital was headed by a Hospital Chief Executive (HCE), with more power and autonomy than the previous Medical Superintendents, and assisted by a number of General Managers in clinical and business functions. The HCEs reported directly to the CE of the HA. Appointment of HCEs went through a nurturing process, assisted by a Management Transformation Implementation Task Force from the Head Office. The first batch of HCEs was appointed in March 1992, and appointments for all other hospitals were largely completed within 3 years. Eighteen of the newly appointed HCEs were senior doctors, reflecting a major effort by the HA to bring new blood into management roles from clinicians within the system. Seventeen other HCEs were former Medical Superintendents, and four HCEs of smaller hospitals came from nursing or allied health backgrounds.7 Internal hospital appointments to new management positions and structural remodelling followed accordingly.

Each clinical department was headed by a Chief of Service, who reported to the HCE, and was assisted by a Departmental Operations Manager (DOM) who would be a senior nurse.8 Each ward was headed by a Ward Manager, also a senior nurse, who reported to the DOM. All nurses working in a clinical department therefore ultimately reported to the Chief of Service as part of an integrated multidisciplinary team, rather than to the Chief Nursing Officer in a hierarchical manner as in the old days. The General Manager (Nursing), who replaced the Chief Nursing Officer position, became part of the HCE’s top management team and no longer held line authority over the DOMs.

Under the new management culture, an inverted pyramid model was advocated, where frontline clinical units were regarded as the most important and better supported by the revamped management structure and capabilities.9 Decentralisation of decision making and participatory management were to be encouraged.

No reform would be successful by merely putting old wine in new bottles. Whatever new mission statement, management positions, systems and processes were put in place would need to be embraced and operationalised by the very people within the system. Given the staff unions’ animosity towards the old regime and suspicion of the new, as reflected in Part 1 of this series,1 winning them over was of utmost importance. An attractive new remuneration package would be a crucial first step. On the other hand, as the Government would ultimately shoulder the HA’s staff costs, consideration should be given to the long-term financial burden and comparability with Civil Service terms.10 Both fronts had to be tackled by the HA.

Under the principle of “what the total cost to Government of running the service would be had all staff been given Civil Service terms”, the approach was taken to divide all staff in the former Government hospital system into pay bands for separate analysis, as the ‘fringe benefits’ differed across bands. An arbitrary snapshot of the situation as at 1 April 1989 was taken to calculate the total cost per staff band, divided by the number of staff members to yield the averaged-out cost per staff member in that band. This would form the basis for constructing the new HA staff terms ‘at comparable cost’.11

The PHA engaged Tower, Perrin, Forster and Crosby to develop a new HA remuneration package (Table 312 13 14 15). To increase its attractiveness, the idea was to change the pension portion into a Provident Fund arrangement that would be invested to generate a higher yield, and to reduce housing and other ‘in kind’ benefits so as to translate more of these into cash. Amid the booming economy at the time, immediate cash, despite the ‘averaging out’ effect, was often more attractive than potential in-kind benefits, and certainly much more flexible for the staff. Whether existing staff would choose to switch over was a matter of individual consideration.

Having settled the new HA terms, another challenge arose concerning the ‘bridging over’ arrangement for Civil Servants in Schedule 1 (ex-Government) hospitals opting into the new terms. Unlike staff employed in Schedule 2 (ex-subvented) hospitals who could simply take out their respective provident fund balances from their previous employers to join the new HA package, those working in Schedule 1 hospitals were mostly on permanent, pensionable terms. The staff unions hoped for some kind of ‘pay-off’ to sever the link with the Civil Service before switching over to the new HA package. This would however incur a large sum of upfront payment from the Government. Moreover, not all staff would work to retirement age and attract a pension. The idea of a deferred pension emerged, namely, one only obtainable upon retirement from the HA and computed using the same methodology of years of service (as Civil Servants) and the last salary drawn before switching over to HA term.16

All existing staff were given an irrevocable choice within 3 years to switch to the new HA terms or remain in their existing terms. Irrespective of terms, all would be subject to the same management on a fair basis within the HA. By the end of the option period, 99.5% of Schedule 2 hospital staff and 58.1% of Schedule 1 hospital staff changed over to the new terms, giving an overall figure of 74.8%. The generally inferior original employment terms in various Schedule 2 hospitals compared with their Civil Service counterparts explained the former’s high conversion rate. Among former Civil Servants in the system, lower-ranking staff bands with less to gain from the new package tended to record fewer switching. Some also suspected that staff on HA terms might have less job security than Civil Servants ‘protected’ by their strong unions.

For clarity, all new hires after the HA took over operation of all public hospitals on 1 December 1991 were only offered the HA terms of employment. Settling staff terms of employment was just a prerequisite, raising staff understanding and performance in the management reform was most important to ensure success. Major resources were thereby committed to support tailored management training courses for senior, middle and frontline clinical staff, equipping them with the concepts and know-how to carry out their new roles. Additional resources were put into the professional training of clinical staff to enhance their competence, job satisfaction and retention. There were also specific training courses for other staff to uplift their performance in areas such as customer service and complaints management. Expansion of management functions also meant introduction of external expertise in a host of non-clinical areas such as information technology (IT), finance, legal, engineering, human resources and other areas of general management.

With improved staff packages, training and professional advancement opportunities, there was an atmosphere of progression and high morale in the new organisation. As a result, staff wastage rate quickly dwindled.

On clinical services, task forces were formed to tackle overcrowding, waiting time, accident and emergency service improvements, nursing services, and more on a territory-wide basis. Clinical Coordinating Committees were formed for each clinical specialty to foster inter-hospital collaboration and service planning to improve system-wide performance in quality and efficiency.

Priority areas of improvement included better inter-hospital cooperation for patient diversion from the most severely overloaded Schedule 1 hospitals to Schedule 2 hospitals, as well as better bed management within each hospital, resulting in drastic reduction in ‘camp beds’. A triage system was implemented in all accident and emergency departments to ensure minimal waiting time for urgent cases. Computerised booking system for Specialist Outpatient Clinics and doubling of effort to increase throughput led to shorter waiting lists and improved access to specialist care. New signages, open counters, upgraded furniture and hospital environment, as well as air-conditioning projects also completely transformed the image of public hospitals.

The Pamela Youde Nethersole Eastern Hospital was completed in 1993, eventually adding more than 1800 beds to the eastern part of Hong Kong Island. In view of the low bed-to-population ratio in northern New Territories, the HA planned and built the North District Hospital, which opened in 1998. There were numerous other additional blocks, extensions, and improvement projects for existing hospitals. Investment in major equipment included the first magnetic resonance imaging machine in the public system installed at Queen Elizabeth Hospital, and additional computerised tomography scanners in major hospitals, etc.

There were hardly any major IT systems in use in public hospitals before the HA, save for very basic ones for payroll and accounting. There was massive investment to revamp these systems and to build many other essential systems for patient registration and appointments, billing and revenue collection, medical records tracing, pharmaceutical management, laboratory results reporting, inventory management, medical equipment management and so on. This original weakness turned out to be a blessing, as it largely obviated the pain of needing to integrate multiple legacy IT systems from different provider organisations, as often encountered overseas when trying to unify data definitions and functionalities, etc. for territory-wide connectivity. The HA also chose to mostly build rather than buy IT systems to maximally fit its own needs and circumstances.

Eventually, the HA embarked on a most comprehensive Clinical Management System to support the work of clinicians and enhance service quality and patient safety. The system became internationally acclaimed in the field of medical informatics, and indeed pride of the organisation. It was clinician-driven from the start, tailored to the clinical workflow, and incorporated advanced features to help doctors and nurses in decision making such as drug allergy and dosage alerts, knowledge support for evidence-based medicine, and so on.17

The HA launched the Patients’ Charter that explicitly listed patients’ rights and responsibilities, with extensive staff communication to change the former ‘we (doctor/nurse) know best’ attitude. Patient feedback on service quality and staff performance was systematically collected. Full-time Patient Relations Officers were employed in hospitals to deal with complaints and suggestions. Patient Resource Centres were set up in hospitals, while an HA InfoWorld was eventually established in the new HA Building to provide a health promotion and patient education platform for the public.

As mentioned above, the Public Complaints Committee incorporating members of the community provided an independent platform for appeals. Partnership with the community was enshrined through the appointment of external members, including patient advocacy organisations, to different levels of governance, a far cry from the closed system of the past.

After adjusting for the resources required to uplift the terms of employment of Schedule 2 hospitals’ staff to be comparable to that of Schedule 1 hospitals, the baseline recurrent budget of the HA for maintaining the same level, scope and volume of services at the beginning of financial year 1992/93 was agreed to be HK$10 301 million.18 There were additional upfront allocations that represented a true increase in investment in the HA to kick-start the reform, including HK$198 million for new projects,19 HK$98 million for new management initiatives,20 HK$90 million for capital projects, and HK$70 million for IT projects.18

Significant improvements in system capacity and efficiency in the HA, underpinned by substantial Government investment, can be seen in Table 4 which compares the full-year effect after the HA’s takeover of management with that 5 years later.21

A number of observations can be made:

1. The number of hospital beds increased more than population growth (15.7% vs 12.6%), improving the number of beds per 1000 population.
2. Growth in inpatient discharges (35.0% general and 32.4% psychiatric) greatly exceeded growth in total bed numbers (15.7%), indicating more active patient management, as also reflected by shortened average lengths of stay for both general (by 7.5%) and psychiatric (by 11.2%) hospitals.
3. The overall occupancy rate improved for general hospitals (from 76.4% to 82.0%), following better utilisation of beds in Schedule 2 hospitals and convalescent hospitals.
4. Severe overcrowding in psychiatric hospitals was reduced (from 95.4% to 89.1% occupancy).
5. An increase in staff productivity is evidenced by the increase in activities (35.0% for general inpatients, 32.4% for psychiatric inpatients, 38.3% for out-patients, 48.3% for accident and emergency attendances, and 41.7% for community nurse visits) exceeding the increase in staff numbers (32.3%). This does not yet reflect the immense improvements in service quality.
6. Taking inflation into account,22 the increase in Government funding to the HA (real growth around 57.8%) and growth in expenditure (real growth around 59.2%) exceeded the increase in staff numbers and activities. As staff costs constituted more than 75% of total expenditure, this mainly reflected the creation of senior posts and general improvement in remuneration.

The HA reform represented a bold social experiment of unprecedented scale in Hong Kong’s history, given that the HA became the second largest employer, after the Civil Service, upon its takeover of all Government and subvented hospitals in one go. The direction of reform followed the then prevalent international trend of new managerialism and corporatisation to free the entity from rigidities and confines of the Civil Service, and went much further than the earlier Housing Authority reform.23 It coincided with a period of economic prosperity during which the Government could afford investing heavily in upgrading staff terms, funding new management initiatives, hospital infrastructure, computerisation, service improvements, and staff training and development.

While Part 1 of this series describes the success story of setting up the new HA,1 the management takeover was when ‘the rubber hits the road’ that could make or break any well-intentioned reform. Led by a visionary Board with high-calibre experts from various fields, the Authority took a pragmatic path by selecting leaders for major executive positions from among influential clinicians within the system with a track record of being reformists who were passionate to change the dysfunctional system of the past.24 The atmosphere was nothing short of a brave new world where the energy of doctors, nurses, allied health staff and administrators was unleashed to learn modern management concepts and methods, and apply them to better the service. This was, needless to say, highly appreciated by the public, to the extent that the private sector felt threatened by a loss of competitiveness in attracting both patients and talented staff.25

While immense success was justified to describe this early period, new issues also emerged. The abolition of the previous Regional Offices level and the emphasis on individual hospital’s autonomy aimed to free up local initiatives and promote internal competition to improve quality and efficiency. However, this also led to reduced cooperation and, arguably, an over-proliferation of management positions (even small hospitals were supposed to have their own HCEs and a full complement of managers). This also meant that the CE had dozens of direct reports, including HCEs and Head Office deputies. As the system evolved, the concept of hospital clustering became increasingly emphasised to streamline management, initially by function and eventually through formal structure. At the hospital level, it remains uncertain to what extent the inverted pyramid model and participative management was fully embraced, depending on the style and preferences of the rather autonomous HCEs. Nevertheless, these were merely minor issues for any major reforms of this complexity, and paled in significance compared to the overall achievement.

As in all such reforms involving tens of thousands of staff, managing the transition was critical. From a historical perspective, the bridging-over arrangement and the new HA terms represented a major victory for staff coupled with the clever design plus political clout of the then HA Board in obtaining extra resources. Such a generous offer had apparently not been repeated since for any other corporatisation exercise of Government functions.

On the other hand, the retained linkage to the Civil Service pay scales and salary point systems also imposed limitation in flexibility when responding to changing circumstances. When the economy turned south and particularly after the Asian Economic Crisis in 1998 when the HA faced budget cuts from the Government, continued hiring on the original terms to cope with the ever-increasing service demand became increasingly untenable. The HA had no choice but to repeatedly create new ranks with less favourable packages, to circumvent the rank-to-rank comparability with the Civil Service, much to the dismay of new hires. The old culture of the ‘iron rice bowl’ expecting job stability and annual salary increments also persisted.

Be that as it may, the experience of turning such a comprehensive reform blueprint into reality with success for such an enormous public system, rendered the HA internationally famous, especially in the healthcare management circles. This was achieved not by simply copying other models, but by integrating best practices from multiple fields and adapting them to the local situation to address Hong Kong’s unique circumstances. An important factor worth mentioning was the cross-disciplinary learning that happened during that period. There were intense interactions among Board members, each with distinct expertise in their own fields, with the senior executives. Conversely, Board members also came to better understand the public healthcare system and its contextual issues. Among the senior executives, there were also vibrant mutual learning between those with clinical and non-clinical disciplines (in management, finance, IT, legal, engineering, etc.) which provided synergistic results.26

Once the newborn organisation was firmly on its feet with a new set of management structure, processes and systems in place, a new phase of reform outgrew the original blueprint. As the appointed senior executives including the CE were predominantly clinician-managers, the subsequent trajectory was heavily influenced by the clinical and public health perspectives, rather than simply system efficiency and customer-focus concerns. The next article (Part 3) describes the emergent philosophy and practice of deepening reform within the HA in the subsequent years.

The author would like to thank the Hospital Authority for permission to use the historical photo of its founding members in this article, as scanned from its Annual Report 1990-1991.

The author declares full responsibility for the accuracy of the content, which does not represent the views of the Hospital Authority. Given the scale of the reform, not all aspects can be covered in detail. Interested readers and scholars are encouraged to consult the Hospital Authority’s publications for further information.

1. Ho W. The Hong Kong Hospital Authority reform: a historical perspective. Part 1: From pre–Hospital Authority era to establishment of the Hospital Authority. Hong Kong Med J 2025;31:508-14.

2. Coopers & Lybrand WD Scott. The Delivery of Medical Services in Hospitals—A Report for the Hong Kong Government. Hong Kong: Hong Kong Government Printer; December 1985.

3. Chung SY. Provisional Hospital Authority Report. Hong Kong: Hong Kong Government Printer; December 1989.

4. Also refer Chung SY, The Hospital Authority of Hong Kong–Part I: From Inception to Reality and its Initial Success. The Medical and Dental Directory of Hong Kong, fifth edition, 1994.

5. Hospital Authority, Hospital Authority Annual Report 1990-1991, pp 5-6. This version of the mission statement was in use until 2009, when the HA Board considerably simplified it.

6. Hospital Authority. Hospital Authority Annual Report 1990-1991, Appendix 2.

7. This is quite different from the picture in the UK, United States or Australia where hospital chiefs are more often non-doctors.

8. Except for radiology and radiotherapy departments where the Department Manager (DM) would be a senior radiographer; pathology departments where the DM would be a senior laboratory technologist; and allied health departments where the DM would be from their own discipline.

9. This puts patient care at the clinical frontline into centre stage at the ‘top’, while the ‘top management team’ should assume a more supportive rather than directive function at the ‘bottom’.

10. The Scott Report proposed the equalisation of staff terms by aiming at a level between that of Schedule 1 and Schedule 2 hospitals, which was vehemently opposed by those working in the former. From the start, therefore, the new HA package was designed to benchmark that of the Civil Servants, which meant an up-front recurrent investment to bring up those working in Schedule 2 hospitals.

11. While all staff of a particular pay band in the Civil Service may be entitled to a host of benefits, not all of them were enjoying all benefits at the same time (eg, government quarters depending on availability), hence the snapshot approach for computing comparable cost.

12. While not explicitly stated, it was implied that retention of the basic salary scales also meant retention of the yearly salary increment practice. In addition, the annual salary adjustments in line with inflation also followed that of the Civil Service.

13. Linking of the Cash Allowance as a percentage of basic salary was under the assumption that in time, the Government would also increase benefits for Civil Servants together with salary increase. However, it turned out that Government was cutting back Civil Service housing benefits a few years down the road. Following a report of the Director of Audit in 1994 that alleged breaching of the comparability principle, the Cash Allowance for new hires was delinked from basic salary.

14. As it played out, the HA Provident Fund Scheme was among the best-performing retirement funds in Hong Kong, both in terms of its administration and investment returns, thanks to the expertise in the Trust Board as well as professionals employed to manage the scheme.

15. With time, the interest portion of mortgages would decrease as will the Home Loan Interest Subsidy Scheme subsidy amount, but the staff member would generally have had pay rise or even promotion, hence could better afford the mortgage. It also meant regenerating available fund to the Scheme to support new applications.

16. Staff unions successfully negotiated using the last drawn salary point as Civil Servant (value updated to the date of retirement from HA) for calculation in the Frozen Pension arrangement. Another option was the Mixed Service Pension arrangement where the staff would retain full pension eligibility for total years served including as HA staff, but receive a reduced Cash Allowance and would not be eligible for the HA Provident Fund. There was also negotiation on the accumulated leave days, resulting in agreement that each staff member was allowed to carry over a maximum of 30 days of ‘sinking leave’.

17. In contrast, many overseas systems are fragmented across hospitals and often structured on the business/financial side while weak on clinical side. The Clinical Management System eventually evolved to become the territory-wide Electronic Health Record (eHR) system that spans the private sector as well.

18. Hospital Authority, Hospital Authority Annual Report 1992-93, p 18. Since HA only took over operation on 1 December 1991, the budget for 1992/93 was the first full-year budget.

19. This amount was for commissioning and opening of new beds and facilities, which would have to be spent with or without the HA’s establishment.

20. Mainly the salaries of all new management positions in the Head Office and hospitals.

21. Hospital Authority. Hospital Authority Annual Reports 1992/93 and 1996/97, Appendices.

22. Inflation figures in Hong Kong were 8.8% (1993), 8.7% (1994), 9.1% (1995) and 6.3% (1996). International Financial Statistics database, International Monetary Fund. Available from: https://data.worldbank.org/indicator/FP.CPI.TOTL.ZG?locations=HK. Accessed 29 Dec 2025. This gives a cumulative inflation of around 37.2% for the period.

23. The Housing Department, which is the executive arm of the Hong Kong Housing Authority, remained within the Civil Service.

24. The first two CEs of HA, Dr EK Yeoh and the author, were active members of the former Government Doctors’ Association, as were Deputy Directors Dr WM Ko 高永文 and Dr Hong Fung 馮康. It is noteworthy that Dr Yeoh, Dr York Chow 周一嶽 (one among the first batch of HCEs) and Dr Ko later became the three successive policy secretaries in the Government (Secretary for Health and Welfare/Secretary for Food and Health) spanning the period 1999 to 2017.

25. The private hospitals began to get their acts together and formed the Hong Kong Private Hospitals Association (HKPHA) in 2000, as well as introduced the UK-based Trent Accreditation system in an effort to improve their own management and quality of service. The author became Chairman of HKPHA in 2018 and thus conversant with its history.

26. Such cross-disciplinary learning seemed to have lost momentum in recent times, as fewer clinicians promoted to management positions pursue formal management courses, while non-clinician executives entirely home-grown within the system may not be able to bring in fresh perspectives and expertise as in the formative years.