A REVIEW OF NEW ZEALAND NEWBORN INTENSIVE CARE SERVICES 1992

Transcription

1 0 A REVIEW OF NEW ZEALAND NEWBORN INTENSIVE CARE SERVICES 1992 Caroline Maskill Department of Health TETARI ORA HEALTH RESEARCH SERVICES

2 A REVIEW OF NEW ZEALAND NEWBORN INTENSIVE CARE SERVICES 1992 Caroline Maskill A report for the Health of Women and Younger People Policy Section Prepared by Health Research Services, DEPARTMENT OF HEALTH, P0 BOX 5013, WELLINGTON April 1992

3 Copyright No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, photocopying, recording or otherwise, without the prior written permission of the Department of Health. Disclaimer This report has been prepared by Caroline Maskill of Health Research Services, Department of Health. Its purpose is to inform discussion and assist in future policy development. The opinions expressed in the report do not necessarily reflect the official views of the Department of Health. Client This project was commissioned by the Health of Women and Younger People Policy Section, Department of Health. It is an internal report, primarily intended for distribution within the Department of Health. It has not been reviewed outside the Department.

4 Contents Summary... 1 Section 1 Introduction Newborn care services Aims of the review Methods Definition of 'neonatal Definitions of low birthweight, prematurity and intrauterine growthretardation Definitions of 'intensive' and 'special' newborn care 'Levels' of newborn care Structure of the report...15 Section 2 Geographic distribution of services, unit size and levels of care Regionalisation Geographic distribution of newborn services in New Zealand Size and levels of care of units visited Geographical access to services...22 Section 3 Profile of newborns and outcomes of intensive care Risk of needing newborn intensive care Admissions to New Zealand newborn units Health outcomes...37 Section 4 Transfers and transport Transfers/transport and regionalisation When to transfer newborns Organising transfers Transport/transfer co-ordinators Transport methods and problems 'Back transport'...46 Section 5 Facilities and staffing Equipment Design of newborn units Diagnostic and transfusion services New technology Staff resources Home care services...59 (iii]

5 Section 6 Quality assurance and information requirements What is quality assurance? New Zealand legislation on newborn care Quality assurance in New Zealand newborn units Research Information requirements...66 Section 7 Financial issues Costs of newborn care in New Zealand Cross boundary charging and diagnosis-related groups (DRGs) Other funding issues in New Zealand Overseas economic analyses...74 Section 8 Newborn services and the new health system Regional health authorities (RHAs) Core services Interim charging Gaining contracts...79 Section 9 Recommendations Data collection Research and monitoring Geographic distribution of newborn services/regionalisation Core services Cross boundary charging Replacing equipment Nurse training, conferences and research New technology...86 Appendix Bibliography [iv)

6 Tables Table 1.1 Table 1.2 Table 1.3 Table 1.4 Table 1.5 Low birthweight categories...11 Definition of neonatal intensive care...11 Definition of neonatal special care...12 Department of Health definition of levels of neonatal care...13 Auckland Area Health Board's classification of levels of neonatalcare Table 2.1 Number of cots in the eight New Zealand hospitals visited...21 Table 3.1 Numbers of infants reported to have been discharged from New Zealand 'neonatal intensive care units', Table 3.2 Comparison of national discharge data and hospitals' own admissionsdata...33 Table 3.3 Primary reason for admission to intermediate nursery: inborn infants, Christchurch Women's Hospital...35 Table 3.4 Reasons for admission to Taranaki Base Hospital Neonatal Unit, Table 3.5 TPN treatment in Dunedin, Table 5.1 Interviewees' comments on newborn unit equipment and itsmaintenance...51 Table 5.2 Comments on the physical design of newborn units...53 Table 5.3 Staff resources of newborn intensive/special care units Table 7.1 Costs of treating infants in newborn units in New Zealand...68 Table 7.2 DRGs used by Auckland Area Health Board for costing treatment of out-of-board infants, Table 7.3 Appendix Appendix Appendix Appendix Appendix Appendix Number of discharges, total days stay and total costs of treating out-of-board infants, Table 1 Definitions of levels of neonatal care...88 Table 2 New Zealand hospitals with neonatal units by unit level and area health board, as stated by area health boards...92 Table 3 Indications for transfer of infants for special care in New Zealand (guidelines accompanying Obstetric Regulations 1975)...93 Table 4 Numbers of liveborn infants by area health boards and birthweight, 1990 (provisional data)...95 Table 5 Percentage of liveborn infants in birthweight categories by area health boards, 1990 (provisional data)...95 Table 6 Numbers and percentages of live birthweight groups for NewZealand, (vi

7 Appendix Table 7 Numbers of neonatal deaths of livebom infants , by birthweight categories...97 Appendix Table 8 Percentage of survivors of liveborn infants , by birthweight categories...97 Appendix Table 9 Quality assurance measures in visited newborn units...98 Appendix Table 10 National Working Party's recommendations for 'management' data relating to infants Figures Figure 2.1a North Island hospitals with levels H-ffl newborn units, 1991 (highest level of care as stated by area health boards)...19 Figure 2.1b South Island hospitals with levels il-ifi newborn units, 1991 (highest level of care as stated by area health boards) Figure 2.2a 300km boundaries of level III newborn units in New Zealand - NorthIsland...24 Figure 2.2b 300km boundaries of level ifi newborn units in New Zealand - SouthIsland Figure 3.1 Percentage of low birthweight live births by area health board of domicile and categories of low birthweight, Figure 3.2 Trends in the rates of low birthweight in New Zealand, Figure 3.3 Trends in the numbers of low birthweight births in New Zealand, Figure 3.4 Trends in the percentage of survivors of liveborn infants , by birthweight categories...38 Figure 3.5 Trends in survival of very low birthweight infants (without lethal abnormalities) born in National Women's Hospital Figure 3.6 Survival rates of very low birthweight infants admitted to Dunedin Newborn Intensive Care Unit and (vi]

8 Summary Section 1: Introduction and definitions Newborn intensive and special care services have played a role in decreasing infant mortality rates. Newborn intensive care is expensive because of the equipment and technology used. New issues in newborn care include: the increasing proportion of low birthweight births, prevention programmes, new technology, ethical and economic considerations and reorganisation of New Zealand health services. There is a variety of definitions of 'intensive' and 'special' care and of levels of care (for example, levels II and III). Section 2: Geographic distribution of services, unit size and levels of care Overseas, regionalisation of services is an accepted method of distributing newborn intensive care resources and is believed to optimise outcomes. Regionalisation of newborn services has been developing in New Zealand over the past two decades. There are five officially recognised level ifi units, although a number of level II units provide some level ifi care. The safety of the latter practice is subject to considerable debate. Some hospitals regularly use more cots for intensive and special care than they are funded for. Most are flexible in the way they use cots for intensive and special care. Area health board staff interviewed differed on the ideal size and minimum throughput of newborn intensive care units. The geographic distribution of intensive care services has developed on the grounds of distance as well as on the size of the population served. Geographic distance is a significant stressor for out-of-town families in terms of separation and the financial costs of accommodation and travel. Area health policies on subsidising these costs are inconsistent throughout the country. (11

9 2 A REVIEW OF NEW ZEALAND NEWBORN INTENSIVE CARE SERVICES 1992 Section 3: Profile of newborns and outcomes of intensive care Conditions that put newborn infants at risk of needing intensive or special care include: prematurity, low birthweight, respiratory distress, jaundice, a diabetic mother, congenital anomalies and other medical problems. About 2-3% of newborn infants need intensive care. Six percent of New Zealand liveborn infants are of low birthweight (<2,500g). About 1% have very low birthweight (<1,500g) and 0.4% are of extremely low biithweight (<1,000g). The proportion of infants with low birthweight in New Zealand has increased over the past two decades, with a significant increase in the rate of extremely low birthweight. Locally collected data on admissions to newborn units are kept in inconsistent ways. Nationally collected discharge and (transfer) data are not consistent with local data and are too unreliable for analysis. A difference in the definition of 'neonatal intensive care' is likely to be one of the causes of the poor data quality. Data on reasons for admission to newborn intensive care, treatment received, and mortality and morbidity are collected comprehensively by some newborn units, but not at all by others. Survival rates of extremely low birthweight infants have improved dramatically over the last decade. Newborn infants who need intensive care have a higher rate of long term morbidity than other infants, although permanent disability is still uncommon. Section 4: Transfers and transport Effective transport systems are essential for regionalised newborn care. Antenatal transfer of mothers is preferable to transporting infants at high-risk, although high-risk situations are often not predictable. Transfers between level ifi centres because of a lack of cots has become less common recently because of the increase in cots in Auckland.

10 SUMMARY 3 Transport systems in New Zealand are adequate, apart from in Christchurch which will soon lose its access to fixed-wing aircraft with the closure of Wigram. Transport equipment needs replacing in Wellington, and Taranaki Base has no transport equipment. Section 5: Facilities and staffing The standard of equipment varies throughout the country. Some units such as National Women's, Christchurch and Wellington have old equipment which needs replacing. Most newborn unit budgets do not have the resources for a planned equipment replacement programme. Equipment maintenance services are very good in most newborn units. The physical design of the newer, purpose-built newborn units generally works well. Older units such as Taranaki Base and Christchurch have inadequate designs (they are currently being replaced or redesigned). Surfactant replacement treatment was available in all the visited newborn units, although special funding arrangements have been necessary. There is very little support for introducing extracorporeal membrane oxygenation (ECMO) into New Zealand in the near future. Diagnostic and transfusion services for newborn units are generally satisfactory. Exceptions are Christchurch, where communication between the newborn unit and off-site laboratory could be improved, and Hastings, where out-of-hours tests are sometimes hard to arrange. Medical and nursing staff shortages exist in some newborn units, particularly Christchurch and Wellington. Home care services are being provided to some extent in all of the level ifi centres except Christchurch.

11 4 A REVIEW OF NEW ZEALAND NEWBORN INTENSIVE CARE SERVICES 1992 Section 6: Quality assurance and information requirements Quality assurance measures are becoming well established in some New Zealand newborn units, whereas others have only limited resources to carry out quality assurance. The lack of computerised databases, personnel for data entry and data analysis, national comparative data and long term follow-up services are major barriers to monitoring local outcomes. The previously recommended national perinatal database and epidemiology unit were strongly supported. Most level ifi hospitals have special in-house training for newborn intensive and special care nurses. However, their opportunities to attend outside conferences and training are limited. All the visited newborn units had their own protocols and/or guidelines for treatment and procedures, but there are no nationally agreed standards for these. Several newborn units have carried out consumer feedback surveys. Level ifi units are conducting neonatal medical research, however very little nursing research is being done. Section 7: Financial issues Information on the costs of newborn intensive care is limited in New Zealand, although financial information systems are being introduced. Cross boundary charging is used to transfer funds to area health boards which treat out-of-board infants in their intensive or special care newborn units. Systems of cross boundary charging have changed several times in recent years. Considerable confusion about the current system exists, particularly among medical and nursing staff. Interviewees were generally not satisfied with the current use of diagnosis-related groups (DRGs) for cross boundary charging. Overseas studies have confirmed the inadequacy of using DRGs in costing newborn care. Interviewees thought that newborn units should have received an increase in funding because of the recent increase in the number and survival of very and extremely low birthweight infants.

12 SUMMARY 5 Interviewees thought that financial support (subsidies) for out-of-town parents, funding for staff training, home care and follow-up and preventive services were important. Overseas studies have shown that the costs of intensive and special care of newborns increase with decreasing birthweight. Cost-benefits are less favourable with decreasing birthweight. Section 8: Newborn services and the new health system The establishment of regional health authorities could potentially have positive or negative effects on newborn care. Regionalisation could be strengthened because of the reduced numbers of RHAs compared with area health boards. However, competition between providers could deregionalise services, as has been reported in the United States. Most interviewees thought that all currently available newborn intensive care services should be core services in the future. Most interviewees did not support core services excluding treatment of infants of certain birthweights or gestational ages. The lack of support was for clinical, practical and ethical reasons. Some interviewees thought that core services should be guidelines only, rather than absolute rules for service provision. Interviewees generally did not support interim charging for infants in intensive and special care. There was some indication that this charging may discourage follow-up checks. Interviewees saw marketing, quality of care and adequate equipment as important in gaining contracts to provide newborn services in the future. Section 9: Recommendations Recommendation 1: Comprehensive data on newborns should be collected in a consistent way by all obstetric and newborn units throughout the country. Recommendation 2: These data should be brought together in a national data base. Recommendation 3: Nationally agreed definitions of newborn special and intensive (and/or level H and level ifi) care should be developed.

13 6 A REVIEW OF NEW ZEALAND NEWBORN INTENSIVE CARE SERVICES 1992 Recommendation 4: Research and monitoring on infants needing special and intensive care services (and other infants) should be undertaken at a national level. Recommendation 5: Research and policy development should be undertaken to investigate the increasing incidence of very low birthweight infants and identify means of reducing the rates of low birthweight infants. Recommendation 6: Recommendation 7: Regionalisation of newborn services should continue to be encouraged. Until accurate comparisons on local outcomes can be made, the current distribution of level ifi services should remain. Recommendation 8: The current practices of some level H units in providing advanced level ffi care should be more widely documented and debated, and a national consensus sought. Recommendation 9: Consideration should be given to increasing the number of newborn intensive and special care cots. Recommendation 10: Because of the rapidly changing nature of newborn intensive care services and improvements in outcomes, the definition of core services should be able to be updated at regular intervals. Recommendation 11: Recommendation 12: Recommendation 13: Recommendation 14: Quality measures should be specified as integral to core newborn services, and therefore could be included in contracts. Facilities, accommodation and travel for families should be considered as essential aspects of core newborn services. Retrieval and back-transport services should be specified as core newborn services. Consideration should be given to include long term followup and home care as core services. Recommendation 15: Serious consideration should be given to changing the current diagnosis-related groups-based cross boundary charging system.

14 SUMMARY 7 Recommendation 16: Resources to replace outdated equipment should be a priority for crown health enterprises and regional health authorities. Recommendation 17: Nursing staff should have improved access to outside training, conferences and research opportunities. Recommendation 18: Adequate funding should be provided for new technology that is proved to be effective in newborn intensive and special care.

15 Section 1 Introduction 1.1 Newborn care services Newborn special and intensive care services are considered to have contributed significantly to decreased infant mortality rates (Budetti and McManus 1982; Corman and Grossman 1985). A previous review of New Zealand neonatal intensive care services concluded that 'newborn special care services, because of their potential for saving life and preventing handicap, should be regarded as an important national health priority' (Maternity Services Committee 1982:5). The 1982 review also recommended that 'special care services for the newborn should be reviewed again after a suitable interval, as part of a review of all services for reproductive health' (ibid.:45). However, the financial cost of intensive care is high. Equipment is expensive and staff need an advanced level of training (Australian Health Ministers' Advisory Council 1991). As resources for neonatal intensive care services are limited, many countries (including New Zealand) have allocated these services on a restricted, regional basis. Since the 1982 review, some issues in neonatal intensive care have changed. For example, the proportion of low birth weight births has increased slightly over the last decade (Morrell 1990), and the proportion of very low birthweight births (under 1500g) has increased markedly (Howie 1990). Prevention programmes have been tested overseas (ibid.) and new technology, such as replacement surfactant and ECMO, has been developed (Australian Health Ministers' Advisory Council 1991). There has also been wide debate on the ethical and economic issues of providing intensive care to very low birth weight infants and those with severe health problems (Clarkson 1983; Kuhse et al. 1988). New Zealand health services have also been reorganised into area health boards, and further reorganisation is planned (Upton 1991). This review was commissioned by the Health of Women and Younger People Policy Section of the Department of Health. Because of the recent advances in newborn care and the 1982 recommendation for further review, the Section considered it timely for a formal review to develop national policy on newborn intensive care services. 1.2 Aims of the review The aims of this review are to: identify current issues and problems of neonatal intensive care services in New Zealand; and suggest improvements in services and solutions to the identified problems. [81

16 INTRODUCTION 9 The focus of the review is on the provision of level ifi (intensive) care, in the context of other newborn and obstetric services. Level 11 (special) care is investigated to a lesser extent. 1.3 Methods A literature review, a review of admission and discharge statistics and interviews with hospital staff were used in this review. Literature review A literature review was carried out on newborn intensive care services. The literature was found through a computer based search from the Department of Health library. The search covered New Zealand and overseas literature from 1985 onwards. In addition, some literature was provided by the Health of Women and Younger People Policy Section. The most relevant literature has been summarised in this report and supports comments made by interviewees and the results of the statistical analysis. Statistical review Some information has already been published on outcomes of neonatal intensive care and was incorporated into the review. For example, issues such as the survival of low birth weight babies and national infant mortality rates have been examined previously (Darlow 1991; Department of Health 1990; Maternity Services Committee 1982). Relevant data from Health Statistical Services were analysed. However, the available data were limited to information on infant mortality and birthweight. Some area health boards keep their own data on newborn admissions to intensive and special care. However, because of the inconsistency of these data, and the lack of accessibility to it within the timeframe for this review, a detailed analysis of these data was beyond the scope of this review. Information from area health boards and other agencies Neonatal intensive care issues were discussed with key people in Auckland, Hamilton, Wellington, Christchurch, Dunedin, Hastings and New Plymouth. Forty-two people

17 10 A REVIEW OF NEW ZEALAND NEWBORN INTENSIVE CARE SERVICES 1992 were interviewed, some in groups, and others individually. At least three staff were interviewed in each centre. They included: paediatricians and nursing staff in level ifi (and some level II) units, community nurses; newborn unit managers; area health board service planners, an ethicist; and parent support group (National Women's Hospital). The review also looked at some area health board and newborn unit reports, for example, service statements and operational/strategic plans. Before presenting the detailed results of the review, it is worth clarifying some of the key terms used. While various definitions exist, the ones adopted by this report are decribed below. 1.4 Definition of 'neonatal' The 'neonatal' period is defined for statistical purposes as being within the first 28 days after birth (National Health Statistics Centre 1990). The 'early' neonatal period is from birth to before the seventh day of life, and the 'late' neonatal period is from seven days, to before 28 days. However, 'neonatal' services are not restricted to looking after infants of less than 28 days old. Many infants who are admitted remain in neonatal care units well past the official 'neonatal' period. Therefore, services which care for neonates and some 'post-neonatal' infants (aged up to one year) will be referred to as 'newborn' services in this report. 1.5 Definitions of low birthweight, prematurity and intrauterine growth retardation One of the main reasons for admission to newborn units is low birthweight. This is commonly, but not always, associated with early gestational age i.e., prematurity or preterm delivery (the latter defined as less than 37 weeks gestation). The World Health Organization (1977) defined low birthweight as being less than 2,500g. More detailed categories are shown in Table 1.1.

18 INTRODUCTION 11 Table 1.1: Low birthweight categories Low birthweight Less than 2,500g Very low birthweight Less than 1,500g Extremely low birthweight Less than 1,000g Source: Adapted from Australian Health Ministers' Advisory Council (1990). Low birthweight occurs because of prematurity (preterm birth) or intrauterine growth retardation, or a combination of both. Preterm birth is 'usually defined as birth occurring before the thirty-seventh week of gestation' or 'less than 37 weeks from the last menstrual period'. Intrauterine growth retardation is low birthweight, but over 37 weeks of gestation. Preterm infants usually require more intensive care than full-term infants who have low birthweights (Morrell 1990). 1.6 Definitions of 'intensive' and 'special' newborn care 'Intensive' care is the highest level of newborn care, and is usually provided in level ifi (or above) units. Neonatal intensive care units generally look after critically ill infants, those who are very premature or very low birthweight and those who have undergone major surgery (see Section 3). The Royal College of Physicians (1988:41) summarised neonatal intensive care as follows: Table 1.2: Definition of neonatal intensive care Neonatal intensive care 1. Babies receiving assisted ventilation (intermittent positive ventilation (IPPV), intermittent mandatory ventilation (IMV), constant positive airway pressure (CPAP)), and in the first 24 hours following its withdrawal. 2. Babies receiving total parenteral nutrition. 3. Cardiorespiratory disease which is unstable, including recurrent apnoea requiring constant attention. 4. Babies who have had major surgery, particularly in the first 24 postoperative hours. 5. Babies of less than 30 weeks' gestation during the first 48 hours after birth. 6. Babies who are having convulsions. 7. Babies transported by the staff of the unit concerned. This would usually be between hospitals, or for special investigations or treatment. 8. Babies undergoing major medical procedures, such as arterial catheterisation, peritoneal dialysis or exchange transfusion. Source: Royal College of Physicians (1988)

19 12 A REVIEW OF NEW ZEALAND NEWBORN INTENSIVE CARE SERVICES 1992 'Special' care is a less intensive form of neonatal care, usually provided in level II units. It involves non-routine care of infants who need medical intervention. The Royal College of Physicians (1988:41) summarised neonatal special care as follows: Table 1.3: Definition of neonatal special care Neonatal special care 1. Babies who require continuous monitoring of respiration or heart rate, or by transcutaneous transducers. 2. Babies who are receiving additional oxygen. 3. Babies who are receiving intravenous glucose and electrolyte solutions. 4. Babies who are being tube fed. 5. Babies who have had minor surgery in the previous 24 hours. 6. Babies with a tracheostomy. 7. Dying babies. 8. Babies who are being barrier nursed. 9. Babies receiving phototherapy. 10. Babies who receive special monitoring (for example frequent glucose or bilirubin estimations). 11. Other babies receiving constant supervision (for example babies whose mothers are drug addicts). 12. Babies receiving antibiotics. 13. Babies with conditions requiring radiological examination or other methods of imaging. Source: Royal College of Physicians (1988) 1.7 'Levels' of newborn care Defining newborn care by the levels or types of care required helps to allocate units within regions or countries, and determine the number of staff needed to look after infants. Different definitions of the various levels of newborn care exist.

20 INTRODUCTION 13 The New Zealand Department of Health's (1991) most recent definitions of the levels of childbirth and newborn services are as follows: Table 1.4: Department of Health definition of levels of neonatal care Level 0 Care provided by general practitioner(s) and midwives for uncomplicated deliveries and well baby care. Level I Service provides care for uncomplicated deliveries and well baby care. Facilities, including a blood bank, are provided for emergency caesarean section and emergency care when necessitated by geographical distance. Level I lacks a combined obstetric/paediatric service and is not a referral centre. Level II Service provides specialist and paediatric care, which includes a limited obstetric and newborn referral centre with limited facilities for newborn intensive care. To be provided by all area health boards. Level Ill A regional specialty service which includes a full range of obstetric and paediatric care, including newborn intensive care, that acts as a major obstetric and newborn referral centre for particular regions and operates a newborn transport service. Level Ill services should be co-ordinated and planned jointly by various regions to provide a nationally unified service for the whole of New Zealand. Should be provided by Auckland, Waikato, Wellington, Canterbury and Otago Area Health Boards. [Level lvi National neonatal surgical and intensive care services for premature infants and serious newborn conditions. Surgical and intensive care facilities should be on the same hospital site and co-ordinated with newborn paediatric medical services. These services are provided by Auckland and Wellington Area Health Boards. Source: Department of Health (1991) These definitions were adopted after extensive consultation with area health boards. Some other recent definitions are provided in Appendix Table 1. Because definitions of levels of newborn units are not always the most useful for planning and staff-mg purposes, some services have developed definitions based on infants' dependency levels. The Auckland Area Health Board (1991:79), in a review of their local neonatal services, proposed a continuous scale of neonatal intensive and special care. This was because they believed 'a clear differentiation between level ifi and level II is often unrealistic in practical situations where there are skilled staff'.

21 14 A REVIEW OFNEWZEALAND NEWBORN INTENSIVE CARE SERVICES 1992 Table 1.5: Auckland Area Health Board's classification of levels of. neonatal care Examples ufiki1lh Pro and Care Needs, Nursing SLafT Implications preterre neonate requiring full ventilation, paralysed pre or post-surgery continuous infusion of medication monitoring of vita] functions every 15 minutes Exchange transfusion planned full ventilation for specific time e.g. for 24 hrs post surgery Fully ventilated - stable Endotzncbial CPAP Nasopbaxyngeal CPAP Hourly feeds, regular monitoring Self respiration some of the time - switch as required preparing for home with respiration assistance 1V treatment Phototherapy l-2hourlyfeeds birth asphyxia just off ventilationlcpap 2-3 or 4 hourly feeds feeders and growcrs close monitoring of temperature antibiotics preparing for home Leveifli 0 4) I: 4) U C 4) -n 0 U U C 4) C p C -= C tn. c - Q. > mothercra.ft 4) -n C 0 -n 0 Nurses : Babies 1: 1 1:1.5 1 :1.2 1 :2.3 1:3 or 4 1:4 1: 5 in Wards Source: Auckland Area Health Board (1991:79)

22 INTRODUCTION 15 As Table 1.5 shows, categorising neonates by the levels of care they need allows neonatal units to allocate the approprate staff using staff:infant ratios. 1.8 Structure of the report The report begins by defining levels of neonatal care and discussing the type of infants who require intensive and special care (Section 1). Sections 2-8 provide the results of the review under the following headings: geographical distribution of neonatal services, unit size and levels of care (Section 2); newborn profile and outcomes of intensive care (Section 3); transfers and transport (Section 4); facilities and staff (Section 5); quality assurance and information requirements (Section 6); financial issues (Section 7); and new born services and the new health system (Section 8). The report concludes with a discussion of the main findings and makes some policy recommendations (Section 9).

23 Section 2 Geographic distribution of services, unit size and levels of care 2.1 Regionalisation The costs and the number of technological advances in the past two to three decades have meant that resources have had to be rationed to efficiently distribute skilled staff and equipment (Darlow et al. 1989). This has been achieved in most western countries by regionalising neonatal intensive and special care. Regionalisation of perinatal health services was considered advantageous in the 1970s. The New Zealand Maternity Services Committee (1982:22) cited the United States Committee on Perinatal Health (1976) definition of regionalisation and the perceived advantages: the development, within a geographic area, of a co-ordinated, cooperative system of maternal and perinatal health care in which, by mutual agreements by hospitals and physicians and based upon population needs, the degree of complexity of maternal and perinatal care each hospital is capable of providing is identified so as to accomplish the following objectives: quality care to all pregnant women and newborns, maximal utilisation of highly trained personnel and intensive care facilities, and assurance of reasonable cost-effectiveness. McCormick (1981) stated that regional networks were important so that medical problems, in mothers and babies from a defined population, could be identified as early as possible. Transfers could then be undertaken to larger centres where 'the most sophisticated of intensive techniques' were available. Overseas, decreased infant mortality rates have been, at least partially, attributed to regionalisation of services. For example, Stahlman (1991) described regionalisation in the United States as being cost effective and responsible for significantly lower perinatal morbidity and mortality levels. Stahiman stated that a regional system was 'medically sound and fiscally responsible as it served to provide the most highly skilled individuals with the most sophisticated (and expensive) technology for high-risk patient care'. McCormick et al. (1985) evaluated a specific programme of regionalised neonatal care in the United States (funded by the Robert Wood Johnson Foundation [RWJFJ) by comparing regions with and without the programme from They found that neonatal mortality rates decreased dramatically in all the regions studied. However, there was no significant difference in the outcomes of the areas with the RWJF programme and those without. The authors attributed this result to the fact that regionalised neonatal care [16]

24 GEOGRAPHIC DISTRIBUTION OF SERVICES, UNIT SIZES AND LEVELS OF CARE 17 had been operating in all the regions during the evaluation period. McCormick et al. found that there were changes in the hospital of delivery so that higher risk deliveries more frequently occurred in regional centres. This suggested that regionalisation had encouraged antenatal risk identification and subsequent transfer of medical management to the higher level perinatal centres. In a cost-benefit analysis of care of very low birthweight infants in Rhode Island, Walker et al. (1985) examined outcomes before and after regionalisation ( and ). The study found that mortality rates for 501-1, 500g infants decreased significantly after regionalisation. Neuro-developmental morbidity rates remained the same. Economic benefits were greater than the costs both before and after regionalisation, but the increase in the number of survivors in the second time period produced higher economic benefits. The study showed that the economic benefits were greatest for infants weighing 1, g. and that there was an economic loss for 501-1,000g infants. Since the time of this study, outcomes for extremely low birthweight infants have improved (see Section 3). In a prospective study, Field et al. (1991) investigated short term outcomes of infants admitted in one year to newborn special and intensive care units in the Trent region of the United Kingdom. They concluded that for infants of less than or equal to 28 weeks gestation, the recognised large intensive care units' provided significantly better survival outcomes than smaller units. They also stated that, for infants of more than 28 weeks gestation, there was no difference in mortality rates between the two types of care. However, Mugford (1991) questioned the interpretation of these results because of the possibility that all medical conditions were not controlled for in the study. For example, infants who were considered too sick for transfer to the regional unit and who stayed in the smaller units to die could have increased the mortality rates of the smaller units. Conversely, the mortality rates in the larger units could have been elevated because of infants having more serious, life threatening conditions than those in smaller units. Also, Field et al. did not study long term morbidity levels. Similar criticisms have been made of the study by Rosenblatt et al. (1985) which looked at perinatal mortality rates in levels 1-ifi obstetric hospitals in New Zealand. Mortality rates in small, rural hospitals were found to be lower than in the larger centres, except for infants of very low birthweight. Apart from low birthweight, medical conditions such as serious malformations which could have affected the infants' survival were not taken into account. In utero transfers were also not taken into account (Lancet 1985). Newborn intensive care units which carried out >500 days of ventilated care per year.

25 18 A REVIEW OF NEW ZEALAND NEWBORN INTENSIVE CARE SERVICES 1992 Rosenblatt et al. concluded that the outcome of the study: probably reflects the cautious antenatal practices of general practitioners and the effective regionalisation of perinatal services in New Zealand... There is no evidence that a satisfactory outcome depends on a minimum number of deliveries (Rosenblatt et al. 1985:429). 2.2 Geographic distribution of newborn services in New Zealand Regionalisation of newborn services has been developing in New Zealand over the past two decades. Figure 2.1 shows the geographic distribution of levels H and ifi newborn services in New Zealand in The information provided by area health boards to the Department of Health relates to the highest level of care offered in each board. The distribution of level ifi services, as stated by area health boards, differs to some extent from that specified by the Department of Health's (1991) service statement and that recommended by the Maternal Services Committee (1982). In addition to newborn intensive care units in the country's five main population centres, several units which are usually regarded as being level II units were described as providing level ifi care. Taranaki Area Health Board stated that it did some level ifi work (long term assisted ventilation) and Hawkes Bay Area Health Board defmed Hastings Hospital as a level In centre. Middlemore was also decnbed as doing level 111 work, although it is not yet a full level ifi unit (a decision has been made that it will be). In addition to the eight hospitals which offer level ifi care, 15 hospitals were stated to offer level H care, and 11 offer level I care. For this review, all level ifi units and the two level H units which were reported as carrying out level ifi work were visited. Staff (paediatricians, nurses and planners) were interviewed about their work and the surrounding issues. I

26 GEOGRAPHIC DISTRIBUTION OF SERVICES, UNIT SIZES AND LEVELS OF CARE 19 Figure 2.1a: North Island hospitals with levels Il-Ill newborn units, 1991 (highest level of care as stated by area health boards) RA bod.aj iu A -4 ab4a t0 o.re * k vo d L&t I1IaL. 64..A.U. 6ø..4 Source: Figure by author from Department of Health data (see Appendix Table 2)

27 20 A REVIEW OF NEW ZEALAND NEWBORN INTENSIVE CARE SERVICES 1992 Figure 2.1b: South Island hospitals with levels li-ill newborn units, 1991 (highest level of care as stated by area health boards) F' Source: Figure by author from Department of Health data (see Appendix Table 2)

28 GEOGRAPHIC DISTRIBUTION OF SERVICES, UNIT SIZES AND LEVELS OF CARE Size and levels of care of units visited The sizes of the units as reported by those interviewed in the present review are shown in Table 2.1. There is a considerable amount of flexibility in the use of cots in most centres. Level III cots can be used for level II infants and vice versa. The number of ventilated cots, the amount of physical space and the number of staff available are the main constraints on the maximum number of infants which can be cared for, rather than the 'official' allocation of cots. The number of special care cots depends on the organisation of perinatal services. Some hospitals have separate nurseries or intermediate care where infants go once they are relatively independent. These cots are not counted as intensive or special care. Other hospitals have rooms within their neonatal units for mothers and infants who are almost ready to go home. The level of care that ordinary post-natal wards provide is also variable. For instance, in Dunedin, the post-natal ward routinely cares for infants who need phototherapy or tube feeding up to three times a day. Table 2.1: Number of cots in the eight New Zealand hospitals visited Number of intensive care Number of (additional) (level Ill) cots special care cots National Women's Middlemore 4 (maximum ventilated) 14 Waikato 11 intensive care 14 special care, 11 intermediate care New Plymouth 3 ventilated 5 (maximum 13) Hastings 2 ventilated 9-20 Wellington 14 (up to 18) 11 Christchurch Women's 6 14 special, 10 intermediate Dunedin The figures presented in Table 2.1 are similar to those collected by a meeting of paediatric and area health board staff (Special and Intensive Care Services for the Newborn in New Zealand 1991). In addition, the meeting stated that there is one intensive care cot each in Palmerston North and Invercargill. Staff in several of the newborn units such as Wellington and Christchurch stated that there were insufficient cots for the demand. Extra cots (over the official allocation) are used by many units when demand is high. Peak occupancy rates of up to 150% were reported by interviewees. However, there are staff, equipment and space constraints to using extra cots. Transfers to other centres and refusing admissions from other area health boards are sometimes necessary. Interviewed staff stated that some level ifi work such as long term ventilation is carried out in Hastings, Taranaki Base and Invercargill (and probably in some other level II

29 A REVIEW OF NEW ZEALAND NEWBORN INTENSIVE CARE SER VICES 1992 centres). There was some concern, mainly among staff in level ifi centres, that this practice may not produce the best outcomes for infants. Other level H and III staff thought that long term ventilation was acceptable and safe in level H units because of the particular paediatricians who were working there, and the fact that higher birthweight and less ill infants are usually kept in level H units. Hastings neonatal unit has a policy of transfering all infants of less than 1000g birthweight. Taranaki Base keeps all infants, regardless of weight, unless they need surgery or other specialised treatment which the local hospital cannot provide. Staff in one of the smaller level III units pointed out that, if level H units in their region do not transfer infants to the regional centre, the regional centre has fewer admissions overall. This situation could mean that staff in small regional centres get inadequate neonatal experience. A neonatal consultant from a larger level III centre also commented that level ifi work by lower level units could be taking away resources from the regional centres. There was considerable debate about the ideal size of level III units. Six to 10 cots was suggested by one interviewee as the minimum viable size of a neonatal intensive care unit, based on outcomes from overseas studies. However, staff in one of the smaller level III units asserted that small units tend to have advantages over larger ones, such as: more opportunities for parents to be involved in infant care, for example, skin to skin contact; - more opportunities for families to be involved in medical decisions; siblings are able to visit more easily; and better continuity of care. It is possible that larger units would also be able to offer these advantages with different management practices. It is also probable that the experience of staff and clinical practice influence outcomes, rather than the size of units themselves. 2.4 Geographical access to services Overseas studies on the regional allocation of neonatal care usually focus on the number of intensive or special care cot numbers needed per 1,000 population or per 1,000 births. Very few have investigated the extent to which geographic distance from neonatal care affects outcomes (mortality and morbidity).

30 GEOGRAPHIC DISTRIBUTION OF SERVICES, UNIT SIZES AND LEVELS OF CARE 23 Mugford (1991:411) points out that: Although it may be the case that care in a regional referral unit can improve the chance of survival for babies of gestations of 28 weeks or less, the problem still remains for those providing the service of how to ensure that all the babies have access to appropriate services, especially in rural areas with dispersed populations. Rosenblatt et al. (1985) suggested that regionalisation has the potential to lead to a concentration of services in large urban areas, with the loss of some small rural obstetric hospitals. In rural areas, this would reduce the availability of lower level services which are essential for the referral of high risk pregnancies to regional centres. In a South Australian study, Davies et al. (1982) examined mortality rates of 389 infants who were transported to an Adelaide neonatal intensive care unit from 1978 to Infants who were transferred from more than 300km away had significantly higher mortality rates than those from closer locations. The authors suggested that colder core temperatures of the infants and possible difficulties in transferring mothers antenatally from such distances result in this poor outcome. Davies et al. (1982:70) stated that 'the transport of babies over distances greater than 300km is peculiar to Australia'. Large distances to level in centres are also involved in the New Zealand situation, although there may be closer access to a lower level centre. Figure 2.2 shows areas within 300km of level ifi centres. Note that road distances can be considerably greater. Additionally, the presence of Cook Strait is a major barrier to transporting infants. However, apart from the extreme north of Northland, all the inhabited areas of New Zealand are within 300km of existing level ifi neonatal centres. Staff from several of the smaller centres visited for the present review stated that isolation was the major factor in them keeping infants who needed level ifi care. A few of those interviewed for the current review suggested that only one level III unit may be necessary for the South Island. As there will only be one regional health authority covering this area, the possibility of one South Island level ifi unit is likely to be considered in the future. If there were only one unit, interviewees expected that this would be in Christchurch. This suggestion was on the basis of the population served, the number of deliveries and/or the relatively low number of admissions to the Dunedin unit. In terms of distance, however, the loss of the Dunedin unit would leave Dunedin itself, over half of Otago and all of Southland more than 300km away from the nearest level ifi unit in Christchurch. Many of those interviewed believed that this scenario would be unacceptable to families for social reasons. Likewise, the loss of the Christchurch unit would leave some of Canterbury and a considerable area of the West Coast further than 300km from a level III unit. Christchurch itself would be on the outer edge of a 300km radius of Wellington.

31 24 A REVIEW OF NEW ZEALAND NEWBORN INTENSIVE CARE SERVICES 1992 Figure 2.2a: 300km boundaries of level.111 newborn units in New Zealand - North Island I / / I / I I I 4 / t I I, I 'I I, 'I I / 1<, F / F I Source: Figure by author

32 GEOGRAPHIC DISTRIBUTION OF SERVICES, UNIT SIZES AND LEVELS OF CARE 25 Figure 2.2b: 300km boundaries of level Ill newborn units in New Zealand- South Island / ' S S S. Source: Figure by author

33 26 A REVIEW OF NEW ZEALAND NEWBORN INTENSIVE CARE SERVICES 1992 Even within a 300km radius, transport can be difficult in New Zealand (see Section 6). Isolation was cited by Taranaki Base unit staff as the reason for carrying out long term assisted ventilation. Neonatal staff in Hastings also said that isolation was the reason they tried to keep as many neonates in Hastings as possible. With the current distribution of neonatal intensive care units, families from out of town were reported by those interviewed to find the situation difficult. For example, the report of the Dunedin Newborn Intensive Care Unit (1992) stated: The financial burden for some families can be severe. Also if parents have left siblings behind in the care of families or friends they usually worry about the separation and feel obliged to leave their sick neonate and return home.... Fathers' jobs are often disrupted though most employers will give holiday time or leave of absence. Staff in Dunedin reported that, at times, Invercargill families with infants in the Dunedin neonatal care unit have resorted to sleeping in cars because of the cost of accommodation. Parents from the Parent Care support group at National Women's Hospital confirmed that stress is high for parents from out of town. Transport costs are high and jobs and income are disrupted. Out-of-town mothers were perceived to be solo parents, in effect, because their partners often have to stay at home to work. Even families who live within the main centres may have access problems if they do not own a car and they have to rely on family and friends to help. The interviewed parents pointed out that those at risk of having a low birthweight infant - the young, Maori and solo mothers - often have less access to the resources they need. Area health board accommodation policies vary. Most provide rooms for mothers (for example in ex-nurses homes) at a cost of around $20 per night. Fathers are often not catered for. Some hospitals provide rooms in the neonatal unit for mothers to look after their babies a few days before going home.