Impact of private funding on access to elective hospital treatment in the regions of England and Wales

Transcription

1 EUROPEAN JOURNAL OF PUBLIC HEALTH 2001; 11: Impact of private funding on access to elective hospital treatment in the regions of England and Wales National records survey BRIAN WILLIAMS, PAMELA WHATMOUGH, JANET MCGILL, LESLEY RUSHTON * Background: The UK National Health Service aims to match access to health care to the level of need and to reduce inequalities in the health of sub-populations. One in ten persons have private medical insurance (PMI). This study describes the impact of private purchasing on access to hospital care in regions according to health need. Method: Details of admissions to NHS hospitals in one year and waiting times were obtained from the government's Hospital Episodes Statistics, and of patients in independent hospitals through weighted time samples of records. Data were combined into two groups, state funded and privately funded patients. The prevalence of limiting longstanding illness and the proportions of individuals covered by PMI in Wales and the eight English health regions were obtained from the General Household Survey. Correlation coefficients were calculated for inter-regional relationships between measures of need, provision of resources and levels of activity. Results: Limiting, longstanding illness was significantly associated with NHS resource levels, NHS hospital activity, and total hospital activity, however funded; and inversely with PMI coverage, waiting times for NHS admission and levels of privately funded activity. Waiting times for admission were positively correlated with PMI coverage. Conclusions: Regionally, NHS resources and activity match need. Private hospital use complements lower levels of NHS service. Private consumption does not distort access according to need but in regions with lower levels of NHS activity those least deprived may make relatively more use of NHS hospitals, thus widening the health gap. Small area studies should explore this.. or the last twenty years in the UK the government funding of health services has been targeted towards those sections of the population with higher levels of morbidity and greater socio-economic deprivation. Over the same period, as the mean duration of hospital stay has fallen and primary and community health care services have been strengthened, there has been a planned reduction in the number of acute care beds in the state's National Health Service (NHS) hospitals. NHS funded hospital services remain free at the point of contact, and there is little or no delay in receiving emergency hospital treatment. However, delays for admission to NHS hospitals for elective treatment can be considerable. At the same time the number of subscribers to private medical insurance rose steadily; by 1998, 11% of the population were covered, two-thirds of these through employment-related subscriptions.' Proportionally more households whose heads were employed in professional or managerial positions were insured (14%) than those * B. Williams 1, P. Whatmough', J McGill 1, L Rushton 2 1 School of Community Health Sciences, University of Nottingham, Queen's Medical Centre, Nottingham, UK 2 MRC Institute for Environment and Health, University of Leicester, UK Correspondence: Professor B. Williams, School of Community Health Sciences, University of Nottingham, Queen's Medical Centre, Nottingham NG7 2UH, United Kingdom, tel , fax Keywords: access, funding, inequality, private medicine whose heads were semi-skilled or unskilled (1%). 2 Thus, the socio-economic groups who generally enjoy better health status are also better equipped to purchase care privately. One in seven of all elective hospital treatments received by residents of England and Wales are privately funded, 81% of these through private medical insurance.'' There are over 0 short-stay independent hospitals in England and Wales, with an average of 50 beds in each. 5 Most of these provide only elective treatments and none provide acute trauma care. Ninety per cent of all privately funded, elective hospital stays occur in independent hospitals, the remainder in sections of NHS hospitals reserved for privately funded patients.- 3 Delays in obtaining privately funded treatment are minimal or non-existent. In general, NHS funding and private funding buy a similar range of elective treatments.^' 7 ' 8 Where access to publicly funded, effective treatment is limited or delayed, those who can pay for it to be provided elsewhere benefit sooner. This unequal access may then aggravate the differences in health status among the various socioeconomic groups in the population. The 1998 report of the Independent Enquiry into Inequalities in Health (Acheson Report), which was commissioned by the Department of Health in England, called for a review of how private practice / NHS

2 Private funding and access to hospital relationships affect access to effective treatments, resource allocation and staff deployment. 9 The study reported here explores how the mix of public and privately funded hospital activity affects access to care at the level of the first geographical tier to which central NHS funding is devolved, namely the region. In the financial year there were eight NHS regions in England. Wales, whose NHS service was, and is funded separately from that of England, can be equated to an English region. The average regional population size is 5.5 millions. We compared the amount and nature of elective hospital care purchased by the state or privately in the eight health regions of England and in Wales in one year. We related the pattern of funding to proxy measures of need for elective treatments. METHOD The method has been described in detail elsewhere. 4 Briefly, demographic, funding and clinical data on patients admitted in sample periods of the financial year were obtained from 215 of the 1 acute independent hospitals with operating departments in England and Wales. 10 ' 11 The whole-year numbers were estimated by weighting the sample data according to sampling duration, the time of year, and non-response, and these estimates were validated as previously described. 4 For NHS hospitals, extracts of the latest available data ( ) were obtained from the Department of Health and Welsh Information Agency's Hospital Episodes Statistics for elective admissions (waiting list and booked admission categories) of non-psychiatric, non-maternity patients. These contained similar data items to those from independent hospitals except that the precise source of funding for private patients was not recorded. Data relating to first consultant episodes (98% of all episodes, and synonymous with admissions) and from the independent hospitals were analysed using SPSS. Data on the private patients in NHS hospitals and independent hospitals were amalgamated, as were the data on NHS-funded patients in NHS and independent hospitals to form two groups, 'private patients' and 'NHS patients'. Although, as in two previous surveys, 7 ' 8 these two sources of data were out of phase by a year, NHS hospital episodes in general and acute specialties rose by only 2% between 1996/1997 and the year of the study, 1997/1998 (NHS Executive, personal communication), so the overall picture was little affected by this adjustment. Two proxy measures of need were used. The percentage of persons in each NHS region and in Wales who, in the 1996 General Household Survey, reported limiting longstanding illness was used to indicate the volume of disabling conditions which might be amenable to elective treatments, such as osteoarthritis in relation to total hip joint replacement. 2 Underlying clinical conditions are not identified in the survey. However, in the annual US National Health Interview Survey the chronic conditions most frequently associated with extended periods of restricted activity include deformities or orthopaedic impairments of the back and lower limbs, arthritis and inter-vertebral disc disorders. 12 Secondly, as a measure of unmet, clinically confirmed need the median waiting times for elective admission to NHS hospitals were derived from Hospital Episodes Statistics data. Two measures of the capacity for treating NHS patients in each region were used. From the relevant statistical reports issued by the NHS Executive we noted for each region the number of beds available in acute specialties and the numbers of consultant grade hospital doctors employed in those acute specialties which accommodate most NHS elective admissions and most private patients, namely anaesthetics, general surgery, neurological surgery, gynaecology, ophthalmology, orthopaedic surgery, otorhinolaryngology, paediatric surgery, plastic surgery, thoracic surgery and urology. 14 The scope of the population of each region for purchasing private treatment was indicated by the percentage of persons who were covered by private medical insurance in 1995, according to the General Household Survey. RESULTS Indicative of need, % of the population of England and Wales reported limiting longstanding illness (regional range 19 26%), and the median waiting time for NHSfunded admission for elective treatment was 46 days (regional range days) (table I). There was a wide range of difference in NHS hospital acute bed supply in the regions ( population) but a smaller difference in the supply of consultant doctors in the relevant acute specialties ( ). However, in the UK hospital consultants who are contracted to work in the NHS may work on a full-time basis (11 sessions per week) or may choose to forego one-eleventh of their salaries in return for being allowed to conduct private hospital practice, working, in effect a ten-session week for the NHS. This, and the fact that some consultants have real part-time contracts with the NHS means that the whole-time equivalent consultant commitment is smaller than the absolute numbers might imply. The regional range of whole-time equivalent consultants was also wide ( ). On average, the proportion of the 11-session working week which was devoted to the NHS was 94% (regional range 90-97%). Almost all consultants who practise privately also work for the NHS. The proportion of the population covered by private medical insurance varied in the regions by nearly fourfold ( %). In an estimated 716,110 residents of England and Wales were admitted to independent hospitals for non-emergency treatment, other than abortion; 669,298 were privately funded, 41,942 were NHS-funded and for 4,870 the source of funding was unknown. There were 4,307,780 elective admissions of NHS-funded residents of England and Wales to NHS hospitals in the year and, excluding abortions, there were 70,512 elective

3 EUROPEANJOURNALOFPUBLICHEALTHVOL.il 2001 NO. 4 admissions of private patients. Thus, across NHS and independent hospitals, in total there were 4,349,7 NHS-funded elective patients (8,484 population) and 739,810 private patients (1,440 ). Private patients purchased 14.5% of all elective treatments (regional range 5-24%). At the regional level the proportion of persons reporting limiting longstanding illness was positively correlated with the supply of NHS acute beds and NHS-funded elective admission rates (even though the differences in the levels of whole-time equivalent consultant-grade medical manpower available did not correspond); and inversely with the levels of the population covered by private medical insurance (table 2). The median waiting time for NHS hospital admission was inversely correlated with the level of limiting, longstanding illness, the level of NHS bed provision, and with the admission rate of NHS-funded patients; and positively correlated with the capacity to purchase treatment through private health insurance. There was only a weak inverse correlation between the regional admission rates of NHS-funded patients and private patients. The total elective admission rate (NHSfunded plus private patients) was positively correlated Table 1 Proxy measures of need, and NHS and private sector hospital capacity and activity, NHS Regional Office areas Need Percent population with limiting longstanding illness (1996) a Median waiting time (days), NHS elective surgical admissions ( ) b Capacity England Northern Anglia South & & & North South & West North Wales England Yorkshire Trent Oxford Thames Thames West Midlands West Wales Not available Not available NHS acute specialty beds population (1997) c NHS consultants per 100,000 ('private practice' specialties), 1996" NHS consultants (wte) ('private practice' specialties), 1996 d Percentage of consultant sessions available to NHS Percentage population with private medical insurance cover (1995 f Activity NHS elective admissions population (excluding abomon) b ' e 8,484 8,4 9,419 7,817 6,217 7,9 7,714 8,317 7,881 10,668 12,874 Private patient admissions population (excluding abortion) 6 1,440 1, ,738 2,248 2,136 1, , All elective admissions, NHS and Private per 100,000 population (excluding abortion) 6 9,924 9,711 10,337 8,697 7,955 9,477 9,850 9,698 8,759 11,969 13,527 Private patient admissions as a percentage of total admissions wte: whole time equivalent a: Office for National, Statistics. Living in Britain. Report of the General Household Survey 1996 b: Department of Health, Hospital Episodes Statistics (see Method). c: Department of Health. Bed availability and occupancy. England financial year Welsh Information Service (personal communication). d. NHS Executive. Hospital Medical and Dental Workforce Census, September 1996 (see text for definition of private practice specialties). e: Admissions of residents of region, wherever admitted.

4 Private funding and access to hospital with the level of limiting longstanding illness, and inversely with the median waiting time for NHS admission. Thus, in relation to need, the presence of the privately funded stream did not distort the overall pattern of access to NHS treatment. The levels of consultant manpower, whether measured as absolute numbers or as whole-time equivalents, did not correlate strongly or very differently from each other with levels of public or private sector activity. In general, however, correlations with levels of NHS service were weaker when the absolute numbers of NHS consultants were used. DISCUSSION In the UK hospital treatment is mainly state funded, but a substantial amount is funded privately, and this is limited only by the ability and willingness of people to pay for it. The proportion of elective, non-emergency hospital treatment purchased privately has remained remarkably constant for nearly twenty years, at around one in seven cases. 3 ' 7 ' 8 In that time there have been periodic surges in state investment in the NHS, aimed at shortening waiting times for hospital admission. In theory, this should have reduced the level of demand for private treatment, but it did not. A the national level it appears that the requisite level of state funding for hospital services can be calculated on the assumption that the cost of meeting one seventh of the entire need for elective treatment will not have to be covered. In terms of its effect on equality of access to care, the indications from this study are that at the regional level the amount and deployment of privately funded treatment has little effect on the pattern of NHS service in relation to need. NHS-funded resources and activities are more abundant in the regions where the levels of morbidity are higher; and, in those regions, when the higher level of need for NHS service intervention is identified, it is delivered more rapidly. However, the longer delays experienced in the regions with better health indices suggest that there the level of NHS service provision may be too low. One indicator of this is the lower percentage of consultant working time devoted to the NHS in those regions. If, as a result, delays then fuel an increased level of demand for privately funded care, in terms of insurance coverage the populations of those regions are among the best equipped to purchase it from the private sector providers. Conceivably, however, the uptake of NHS services in the regions with the more favourable health indices is relatively greater among the same socio-economic groups who use the private sector. The available data were not amenable to exploring this. However, if that is so, the more deprived sections of the population of those regions may be doubly disadvantaged. Table 2 Relationships between need, capacity and activity, eight NHS Regional Office Areas, England; and Wales (Pearson correlation coefficient, two-tailed) Percentage with limiting longstanding illness -0.81' 0.88* * 0.85* -0.70* 0.75* NHS median waiting time (days) -0.91" -0.76* -0.82* 0.80* -0.86* * NHS acute beds per 100, * 0.95* -0.68* NHS consultants NHS wte consultants Percentage of population with PMI cover NHS elective admissions ** Private patient admissions ** * All elective admissions NHS and private per 100,000 * Correlation significant at the 0.05 level ** Correlation significant at the 0 01 level

5 EUROPEANJOURNALOFPUBLlCHEALTHVOL.il 2001 NO. 4 Only the broadest indicators of need from some of the most readily accessible, routine data sources were used for this region-based study. The impact of private purchasing on access to care and on health inequality is, of course, better studied at a local level, where more precise and relevant measures of need, socio-economic potential and service utilization may be employed. The project was supported by grants from the Association of British Insurers (AB1) and the Independent Healthcare Association (1HA). We thank the following for supplying information and/or advice: officers of the HES section, Statistics Division, Department of Health, Welsh Health Information Services, IBM UK Global Services, ACT Medisys Ltd, Streets Heaver Healthcare Computing, Tim Baker of Norwich Union Healthcare, and Steve Gameren of Nuffield Hospitals; the information officers of individual hospitals and groups; and the Acute Hospitals Board of the Independent Healthcare Association, the Chief Executives of independent hospital groups and of individual hospitals, for such a high degree of co-operation. 1 Laing W. UK Private Medical Insurance London: Laing & Buisson, Office for National Statistics. Living in Britain. Report of the General Household Survey, London: Office for National Statistics, Williams B, Whatmough P, McGill J, Rushton L. Contribution of the private sector to elective hospital treatment in England and Wales, : national records survey. BMJ 2000;320: Williams B, Whatmough P, McGill J, Rushton L. Patients and procedures in short stay independent hospitals in England and Wales, J Public Health Med 2000,: Directory of Independent Hospitals and Health Services. London: Financial Times Healthcare, Independent Hospitals Association, Williams B. Utilisation of National Health Service hospitals in England by private patients Health Trends 1997, Nicholl JP, Thomas KJ, Williams BT, Knowelden J. The contribution of the private sector to elective surgery in England and Wales. Lancet 1984;ii: Nicholl JP, Beeby NR Williams BT. The role of the private sector in elective surgery in England and Wales, BMJ 1989; Report of the independent enquiry into inequalities and health. London: HM Stationery Office, Independent Healthcare Association. Directory of independent hospitals and health services, London: Churchill Livingstone, Fitzhugh W. The Fitzhugh Directory: independent healthcare and long term care. Financial information, London: HCIS, National Center for Health Statistics. Prevalence of selected chronic conditions: United States, Vital and Health Statistics Series 10, No DHHS Publication No MD: Hyattsville, US Department of Health and Human Services, Department of Health. Bed availability for England. Financial year Leeds: NHS Executive, Department of Health. NHS hospital, public health and community health service. Medical and Dental Workforce Census, England, at 30 September, Detailed results. Leeds: NHS Executive, Received March 2000, accepted 8 September 2000