Family Practice Vol. 18, No. 2, 141-148
© Oxford University Press 2001
An inner city GP unit versus conventional care for elderly patients: prospective comparison of health functioning, use of services and patient satisfaction
Department of R&D, Kensington & Chelsea and Westminster Health Authority and
a Department of Psychiatry and Behavioural Sciences, Royal Free and University College Medical School, London, UK.
Natalie Boston, Senior Researcher in Mental Health Services, Department of Public Health (R&D), Kensington & Chelsea and Westminster Health Authority, 50 Eastbourne Terrace, London W2 6LX, UK.
Boston NK, Boynton PM and Hood S. An inner city GP unit versus conventional care for elderly patients: prospective comparison of health functioning, use of services and patient satisfaction. Family Practice 2001; 18: 141–148.
Received 2 May 2000; Revised 23 August 2000; Accepted 30 October 2000.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Background. GP units are generally nurse-led wards, where GPs have direct admitting rights and retain clinical responsibility for their patients. While GP-led wards are not new, they are relatively uncommon in urban areas. In addition, there has been little comparative evaluation of this type of service.
Objectives. The aim of the present study was to compare patients admitted to an inner city GP unit with comparable patients in conventional care (e.g. district nursing, nursing/residential homes, acute care of the elderly wards) in terms of mental and physical functioning, use of health and social services and patient satisfaction.
Methods. Study group patients were those admitted to the GP unit; comparison group patients were identified by GP practices or conventional services who had agreed to participate in the study. Suitable patients were aged 65 years or over and fitted the eligibility criteria for the GP unit. Patients were interviewed at three time points: admission to either the GP unit or conventional care, and at 1 and 3 months after admission. Baseline comparability was assessed by demographic and medical data, cognitive function, mental state, social support, use of health and social services, and mental and physical functioning (SF-12). Mental and physical functioning and use of health and social services were compared between the groups over time. Patient satisfaction with their care was also compared between groups.
Results. Change in the mental and physical functioning between patients on the GP unit (n = 67) and those in conventional care (n = 60) did not differ when the groups were compared at any of the three time points. However, the mental function of patients in the GP unit significantly improved between admission and 1 month after admission (P < 0.05). This effect was not sustained at 3 months after admission. GP unit patients were consistently more positive about the care they received than patients receiving conventional care; this included communication and information, staff, care and the facilities. Both groups of patients were high users of health and social services, with similar patterns of use in both groups, which did not alter over time.
Conclusions. Patients who received care on the GP unit experienced a similar physical outcome to patients in conventional settings; however, they appeared to enjoy a short-term improvement in mental functioning and were consistently more positive about the quality of their care. This study has important policy implications with regard to planning future intermediate care services and will be of particular interest to health service planners and those responsible for clinical governance.
Keywords. Elderly care, intermediate care, satisfaction.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
GP units are in-patient wards to which GPs have direct admitting rights, and are directly involved in the care of their patients. Wards are generally nurse-led, with provision for a range of multidisciplinary services (such as physiotherapy) but usually without specialist input. While GP units are relatively common in rural areas where district general hospitals (DGHs) may not be easily accessible, few are located within inner cities. London has had only three examples of GP units in recent years—one of which has been generally regarded as a success,1 one which was closed due to low bed occupancy and high staff turnover,2,3 and the Phoenix ward GP unit based at St Charles community hospital which opened in 1995, and is the subject of this paper.
Several factors specific to inner city areas have led to increased interest in intermediate care (e.g. urban GP units) in recent years, which has been suggested as an alternative to care in hospitals for suitable patients,4 particularly for those who are elderly. These include: the shift towards a ‘primary care-led NHS’, including the specific recommendations of the 1992 Tomlinson report5 which concluded that primary care in London was under-resourced at the expense of secondary services; older patients in London are more likely to live in poverty;6 a high proportion of older people live alone compared with younger age groups;7 elderly patients are less likely to gain access to certain health care services and have poorer outcome following utilization of these services;8,9 and GPs in London have documented difficulty in having acute referrals accepted, especially where patients are elderly.10 In addition, a proportion of elderly patients in acute beds are located inappropriately —because of a need for non-acute health services or for nursing home care.11
However, there has been little rigorous comparative evaluation of GP units in the NHS. Descriptive studies have provided some data on activity of GP units, but there have been no published studies examining the extent to which these compare with conventional services. This paper forms part of a wider 2-year prospective evaluation of a 15-bed GP unit (Phoenix ward) in inner London,12 and reports the results of a comparative study of the ward. Phoenix ward was established as a result of the Tomlinson recommendations with the aim of bridging acute and primary services; patients are referred via their GP for treatment, rehabilitation and terminal care. This study compared patients admitted to the GP unit with comparable patients admitted to conventional care in terms of health functioning, health and social service use, and patient satisfaction.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
This was a prospective non-randomized comparative study of elderly patients admitted to a GP unit and comparable patients admitted to conventional services in an inner London health authority between February 1998 and March 1999. Study group patients were those admitted to the GP unit; comparison group patients were those who fitted the criteria for the GP unit (see below) who were admitted to conventional forms of care identified in a map of services13 (including local community schemes, district nursing and acute care of the elderly and general medical wards). All GP practices within the health authority were contacted by researchers and asked to identify suitable comparison patients to them. Following low rates of patient referrals to the study by participating GPs, conventional services were contacted directly by researchers to identify patients for the study. Suitable comparison patients in both groups were contacted by researchers and asked to consent to participate in the study.
Patients were accepted onto the study if they were aged 65 years or over, gave informed consent to participate, were resident within the health authority area and met the criteria for admission to the GP unit (patients requiring respite care, rehabilitation, convalescence and non-acute treatment or investigation and who were not suffering from dementia or a psychiatric illness). Patients who were demented, suffered from an acute psychiatric illness, had an infectious condition or required specialist medical treatment were excluded from the study (these were exclusion criteria set by the GP unit for admission to the ward; adopting these criteria ensured patients referred to the study were representative in both groups).
Patients were interviewed at three time points: 5–8 days after admission, 1 month after admission and 3 months after admission. Baseline comparability between groups was assessed by demographic and medical data, cognitive function (Abbreviated Mental Test),14 mental state (Philadelphia Morale Scale)15 and mental and physical functioning (respective component summary scores of the SF-12).16 Social support was measured in terms of quality and quantity of support (i.e. number of relatives or friends patients seen in a month, number of people with whom they could talk frankly, and how much of the time they had someone to listen to and support them and perform practical chores if they were unable to do them themselves). This scale was developed from the well-validated measure used in the Medical Outcomes Social Support Survey17 (originally consisting of 20 items). Change in physical and mental functioning over time was assessed by the SF-12. Patient satisfaction and perceptions of care were assessed at the second interview, using a modified version of a measure used in the National Survey of Hospital Patients.18 Use of health and social services was measured at all three interviews using patient recall.
Following a pilot study that assessed inter-rater reliability, all data was collected onto standardized forms by one of two researchers (NB and PB), and subsequently analysed in SPSS. Means were compared using t-tests, and differences in proportion were assessed by chi-squares (statistical significance was inferred when the P-value was <0.05). Multivariate analysis adjusting for co-morbidity was conducted to determine whether variation in patient satisfaction could be explained by type of care (e.g. GP unit versus conventional care) or type/severity of illness.
Subgroup analysis
To validate data derived from comparing in-patients on the GP unit with patients receiving both in-patient and out-patient comparable conventional care, subgroup analysis was conducted to compare GP unit patients and conventional care hospital in-patients alone.
Sample size calculations
In order to detect mean score differences of at least six points in terms of patient functioning (SF-12), a minimum of 50 patients per group were required (assuming 80% power, 5% significance and a mean score of 40 in the group receiving conventional care with an SD of 10).
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
A total of 228 admissions (172 patients) were made to the GP unit during the study period (1 February 1998– 3 March 1999), of whom 92% were aged over 65 years of age (range 40–98 years). Sixty-six (out of 172) patients were excluded on the basis of non-consent, confusion, they were too ill or were unavailable when researchers visited the ward, or they did not fit the study criteria. A total of 155 comparison group patients were referred to the study, of whom 81 were found to be suitable to take part. Sixty-seven patients (response rate 63%) were recruited to the study from the GP unit and 60 from conventional care services (patients in the comparison group were fairly equally distributed between community and hospital services: 55% versus 45%, respectively). At entry to the study, there were no differences between patients in either the GP unit or conventional care along any of the dimensions measured (Table 1
), with two exceptions: patients receiving conventional care suffered more breathing difficulties including asthma and bronchitis (chi square = 3.9, P = 0.05) (Table 1) and GP unit patients were more likely to have seen their GP prior to admission: 92.5% on the GP unit compared with 80% in conventional care (chi square = 7.9, P = 0.005) (Table 2).
|
|
Mental and physical functioning of GP unit patients were similar to those of conventional care patients at all three time points. Additionally, physical functioning did not change over time for patients in either group. However, mental functioning of GP unit patients improved between entry to the study and 1 month later (P < 0.05). By 3 months, however, this effect had disappeared (Table 2
).
Use of health and social services over time was also similar for both groups. GP unit patients, however, were more likely to have seen their GP than patients in conventional care 3 months after admission: 46.3% on the GP unit compared with 28.3% in conventional care (chi square = 5.4, P < 0.05) (Table 2).
GP unit patients were consistently more positive about their care than patients in conventional services, especially around areas of communication, information, staff, assistance/care and the facilities (Table 3). GP unit patients felt more informed about their care than those receiving conventional services, for example 57.1% of GP unit patients were given some form of written information about their routine on admission compared with 10% of patients receiving conventional care (however, there were no differences in the number of people who received verbal information: 42% of GP unit patients compared with 49% of patients receiving comparable care). GP unit patients were also more satisfied with the amount of information that was given to their friends/ relatives about their condition to facilitate their recovery compared with patients in conventional care (73% compared with 51.9% in conventional care) and felt more involved in decisions about their care (87.2% on the GP unit compared with 65.9% in conventional care). Patients on the GP unit were twice as likely to say that staff on the ward wore name badges (86.8% compared with 40.6% in conventional care). Patients receiving conventional care were more likely to think that the doctors/nurses were withholding information from them (29.3% compared with 7.9% of GP unit patients), and that staff talked in front of them as if they were not there (26.8% compared with 4% of GP unit patients).
|
Patients on the GP unit were also more satisfied with the staff and facilities. For example, 72% of patients found the doctors, and 84.3% found the nurses helpful (compared with 50 and 40.5% of conventional care patients, respectively), and 82.7% of GP unit patients were very confident in the ability of the staff, compared with 57.5% in conventional care. Similarly, GP unit patients were more likely to say that staff were available when they were needed, for example 90.5% of GP unit patients always got help with bathing when they needed it, and 97.4% said the nursing care was as good at night as during the day (compared with 65.5 and 25%, respectively of patients receiving conventional care). GP unit patients were also more likely to report that the food on the ward met their dietary requirements (76.9% compared to 53.3% of conventional care patients). While more patients on the GP unit were satisfied with the cleanliness of their ward or room (81.6% compared with 44.8% in conventional care), more patients receiving conventional care did not share a room throughout most of their care (55.6% compared with 17% of patients on the GP unit). Finally, a higher number of patients on the GP unit (94%) said they would recommend their care to a friend compared with patients receiving conventional care (79.5%).
When co-morbidity was adjusted for [95% confidence interval (CI)], multivariate analysis revealed that GP unit patients retained higher levels of patient satisfaction than those in conventional care, suggesting that patient perceptions around quality of care were related to use of the GP unit and not morbidity.
Subgroup analysis
When in-patients on the GP unit (n = 67) were compared with conventional care hospital in-patients alone (n = 27), some differences were found between the two groups at entry to the study. While females made up almost three-quarters of the study group, sex was equally distributed in the in-patient comparison group (P < 0.05), which also constituted significantly fewer white patients and more Asians at baseline (P < 0.05). Additionally, in-patients in the comparison group suffered more co-morbid conditions: 42.3% suffered from breathing difficulties (including asthma and bronchitis) compared with 21.2% in the GP unit (P < 0.05), 28% suffered from a stroke compared with 10.9% in the GP unit (P < 0.05) and 23.1% suffered from kidney problems compared with 7.7% in the GP unit group (P < 0.05). However, no differences were found between groups in terms of physical and mental health functioning as measured by the SF-12 at any of the three time points.
Due to the fact that a high proportion of in-patient comparison patients were still in hospital 1 month after admission to the study, comparisons with other patients were influenced. As a result, the in-patient comparison group appeared to make considerably less use of community services compared with GP unit patients [fewer patients used the home help facility both on entry to the study (P = 0.05) and 1 month later (P < 0.05), and they made less use of district nursing services (P < 0.05), meals on wheels (P < 0.05) and day centres (P < 0.05) at 1 month after entry to the study].
Similar statistically significant results were obtained when the two groups were compared in terms of how positive patients were about the quality of their care (five items became non-significant: belief that doctors/nurses were withholding information, family/friends given all information needed to enable patients' recovery, very confident in ability of staff, doctors very helpful and room/ward very clean). Four items found not to be significant in the main analysis now reached statistical significance: 73.1% in the GP unit group thought it was very easy to find someone on the hospital staff to talk to about concerns about their condition/treatment compared with 36.8% in the in-patient comparison group (P < 0.01); 40.4% of GP unit patients were given medication to help with their pain without asking compared with 26.7% of the in-patient comparison group (P < 0.05); a higher number of comparison in-patients had a complaint about poor/inefficient hospital administration (15% compared with 1.9% on the GP unit, P < 0.05); and a greater number of comparison in-patients perceived that they had one nurse in charge of their care (40% compared with 5.8% in the GP unit, P < 0.01).
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
This prospective, comparative study of an inner city intermediate care unit has shown that in terms of patient functioning, levels of care are at least as good as those in conventional services. Further, GP unit patients experience a short-term improvement in mental functioning and consistently higher levels of satisfaction with their care. Taken together, these results suggest that perceived better quality services (described by patients in the intermediate care group above) may lead to increased mental functioning in the form of improved feelings of well-being. Good-quality intermediate care therefore has a clear and important role within the current context of services for physically frail elderly patients.
Both groups were high users of health and social services, in line with national reports of service use.19 Patients on the GP unit were more likely to have seen their GP than those receiving comparable services both prior to admission and 3 months later. This is probably due to the role of GPs in the referral process to the unit, and GPs may make more follow-up visits as a result of their responsibility for their patients during their stay on the GP ward. Use of services remained high in both groups throughout the study period even when most patients had been discharged from care. Neither the GP unit nor conventional care therefore appear to lead to even a short-term reduction in service use.
Randomization deliberately was not chosen as a design for this study as this would have removed the role of patient and GP choice in the admission of patients to the ward, and affected ward occupancy (not reported in this paper).12 Undertaking a randomized controlled trial may have reduced the ‘generalizability’ of the results, for example outcomes in the present study may be highly dependent on the characteristics of the providers of care, the setting and the patients (e.g. GPs on the GP unit could be viewed as having a vested interest in the success of the ward).20
Despite methodological difficulties of comparison with conventional care, we are confident that the non-randomized design that was used produced representative groups of patients. Overall, there were no differences between the study and comparison groups on any of the dimensions we measured, and the sample size was big enough to have detected any differences at baseline. Similarly, patients in both groups were representative of all those referred to the study regardless of whether they participated.
Two main methodological problems were encountered with the present study. First, this study compares in-patients on the GP unit with patients receiving both in-patient and out-patient comparable conventional care. As a result of the heterogeneity of this latter group, subgroup analysis was conducted to investigate whether any differences occurred between GP unit patients and conventional care hospital in-patients alone. Due to the sample size, caution should be exercized when drawing conclusions; however, patients' perceptions of quality of care on the GP unit remained more favourable compared with conventional services for elderly patients. This was supported further by anecdotal statements made by patients to researchers during the course of the study (not reported here). Secondly, some differences in case mix were encountered. For example, although we achieved a highly satisfactory convergence between GP unit and conventional care patients in terms of mental and physical functioning, some key differences in co-morbid conditions were revealed. Thus those satisfaction criteria potentially related to morbidity (e.g. beliefs about doctors/nurses withholding information, nurse availability and help with bathing) may not be the result of more successful communication or better care on the GP unit but rather be the consequence of more serious or life-threatening disease in the conventional care group. However, multivariate analysis between patients on the GP unit and in conventional care determined that variation in patient satisfaction was related to the type of care and not morbidity within the comparison group (the sample size was too small to conduct this analysis with subgroup data).
This study did not examine the cost of the GP unit compared with conventional care; however, a cost-effectiveness study undertaken in 199521 found that the GP unit would be economically viable compared with nearby acute care of the elderly wards if it maintained a minimum occupancy of 80%.
The present study involved the systematic evaluation of elderly patients' experiences with their care using a measure designed to incorporate detailed and specific questions about in-patient care. Taking into account the methodological difficulties inherent in the design of this study, this paper nonetheless provides useful insight into the strength of intermediate care. For example, whilst one might expect those in intermediate (GP unit) care to receive more information and feel generally more involved in their care compared with those receiving community services, arguably patients in all forms of care available to the elderly ought to feel similarly included. Both the Government and the General Practitioner Committee currently are committed to exploring new ways of working within primary care (of which intermediate care is one example), and these findings emphasize some aspects where high-quality care is achieved in an intermediate care facility in inner London—an area where quality of care for elderly patients remains an ongoing area of concern. This study has important policy implications with regard to planning future intermediate care services and will be of particular interest to health service planners and those responsible for clinical governance.
Given the current emphasis on increasing intermediate care facilities in inner London,22 the development of GP units may become more feasible. This study presents evidence that GP units may provide a viable alternative to conventional care in terms of patient experience. Importantly, this study highlights important shortfalls in the quality of conventional care for elderly patients, by no means a new area of concern.23 The consistently low levels of satisfaction with care reported by patients in conventional services are particularly troubling given the relatively positive experiences of GP unit patients—in other words, poor quality of care need not be inevitable. Further work needs to be undertaken to determine those elements of GP unit care which contributed to better results for patients, and to devise ways to generalize this success into mainstream services.
|
Acknowledgments |
|---|
The authors would like to thank the National R&D Programme: Primary/Secondary Care Interface for funding the study, staff and patients on Phoenix ward GP unit and all conventional services which took part in the study, and Angela Crook at Kensington & Chelsea and Westminster Health Authority for her advice on statistics.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1 Henderson G, Tarpey M. Assessing the Lambeth Community Care Centre. London: LCCC, 1990.
2 Victor CR. The Re-evaluation of the Paddington and North Kensington Health Authority GP Community Hospital. London: KCW HCA, 1988.
3 Victor CR. End of an experiment: report from an inner city community hospital. J R Coll Gen Pract 1988; 38: 407–410.[Web of Science][Medline]
4
Coast J, Inglis J, Frankel S. Alternatives to hospital care: what are they and who should decide? Br Med J 1996; 312: 162–166.
5 Tomlinson B. Report of the Inquiry into London's Health Service, Medical Education and Research (Tomlinson Report). London: HMSO, 1992.
6 Hills J. Income and Wealth: The Latest Evidence. York: Joseph Rowntree Foundation, 1998.
7 Office for National Statistics. General Household Survey 1996–7. London: HMSO, 1998.
8 Acheson D. Independent Inquiry into Inequalities in Health. London: HMSO, 1998.
9 The Royal College of Physicians. Ensuring Equity and Quality of Care for Elderly People. London: The Royal College of Physicians, 1994.
10 Jenkins C, Bartholomew J, Gelder F, Morrell D. Arranging hospital admission for acutely ill patients: problems encountered by general practitioners. Br J Gen Pract 1994; 44: 251–254.[Web of Science][Medline]
11 Victor C, Nazareth B, Hudson M, Fulop N. The inappropriate use of acute hospital beds in an inner London district health authority. Health Trends 1993; 25: 94–97.[Medline]
12 Boston N, Boynton P, Hood S. Evaluation of the St Charles Hospital GP Unit (Phoenix Ward): An Urban Community Hospital in Inner London. London: KCW HA, 1999.
13 Boston N. Map of Services for the Physically Frail Elderly in Kensington & Chelsea and Westminster. London: KCW HA, 1998.
14
Hodkinson HM. Evaluation of a mental test score for assessment of mental impairment in the elderly. Age and Ageing 1972; 1: 233–238.
15 Lawton MP. The Philadelphia geriatric centre morale scale: a revision. J Gerontol 1975; 30: 85–89.[Abstract]
16 Ware JE, Kosinski M, Keller SD. A 12-item short-form health survey: construction of scales and preliminary tests of reliability and validity. Med Care 1996; 34: 220–233.[Web of Science][Medline]
17 Sherbourne CD, Stewart AL. The MOS Social Support Survey. Soc Sci Med 1991; 32: 705–714.
18
Bruster S, Jarman B, Bosanquet N, Weston D, Erens R, Delbanco, TL. National survey of hospital patients. Br Med J 1994; 309: 1542–1546.
19
Victor CR, Higginson I. Effectiveness of care for older people: a review. Qual Health Care 1994; 3: 210–216.
20
Black N. Why we need observational studies to evaluate the effectiveness of health care. Br Med J 1996; 312: 1215–1218.
21 London Health Economics Consortium. An Economic Evaluation of the Phoenix Ward GP Unit. London: KCW HA, 1995.
22 Department of Health. Shaping the Future NHS: Long-term Planning for Hospitals and Related Services. Consultation document on the findings of the National Beds Inquiry. London: Department of Health, 2000.
23 Health Advisory Service. Not Because They are Old. London: HAS 2000, 1998.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  P. M Boynton, G. W Wood, and T. Greenhalgh Reaching beyond the white middle classes BMJ, June 12, 2004; 328(7453): 1433 - 1436. [Full Text] [PDF]
|
|
|
|
 |
|
 A. Round, T. Crabb, K. Buckingham, R. Mejzner, V. Pearce, R. Ayres, C. Weeks, and W. Hamilton Six month outcomes after emergency admission of elderly patients to a community or a district general hospital Fam. Pract., April 1, 2004; 21(2): 173 - 179. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||