Injury, Int. J. Care Injured 45 (2014) 902–909
Contents lists available at ScienceDirect
Injury journal homepage: www.elsevier.com/locate/injury
Assessment of quality of life and functional outcome in patients sustaining moderate and major trauma: A multicentre, prospective cohort study T.H. Rainer a,b,*, J.H.H. Yeung a,b, S.K.C. Cheung a, Y.K.Y. Yuen a, W.S. Poon c, H.F. Ho d, C.W. Kam e, G.N. Cattermole f, A. Chang d, F.L. So e, C.A. Graham a,b a
Accident and Emergency Medicine Academic Unit, Chinese University of Hong Kong, Hong Kong Trauma & Emergency Centre, Prince of Wales Hospital, Hong Kong c Division of Neurosurgery, Department of Surgery, Chinese University of Hong Kong, Hong Kong d Accident and Emergency Department, Queen Elizabeth Hospital, Hong Kong e Accident and Emergency Department, Tuen Mun Hospital, Hong Kong f Emergency Department, Princess Royal University Hospital, Orpington, UK b
A R T I C L E I N F O
A B S T R A C T
Article history: Accepted 10 November 2013
Background: Trauma care systems aim to reduce both death and disability, yet there is little data on posttrauma health status and functional outcome. Objectives: To evaluate baseline, discharge, six month and 12 month post-trauma quality of life, functional outcome and predictors of quality of life in Hong Kong. Methods: Multicentre, prospective cohort study using data from the trauma registries of three regional trauma centres in Hong Kong. Trauma patients with an ISS 9 and aged 18 years were included. The main outcome measures were the physical component summary (PCS) score and mental component summary (MCS) scores of the Short-Form 36 (SF36) for health status, and the extended Glasgow Outcome Scale (GOSE) for functional outcome. Results: Between 1 January 2010 and 31 September 2010, 400 patients (mean age 53.3 years; range 18– 106; 69.5% male) were recruited to the study. There were no statistically significant differences in baseline characteristics between responders (N = 177) and surviving non-responders (N = 163). However, there were significant differences between these groups and the group of patients who died (N = 60). Only 16/400 (4%) cases reported a GOSE 7. 62/400 (15.5%) responders reached the HK population norm for PCS. 125/400 (31%) responders reached the HK population norm for MCS. If nonresponders had similar outcomes to responders, then the percentages for GOSE 7 would rise from 4% to 8%, for PCS from 15.5% to 30%, and for MCS from 31% to 60%. Univariate analysis showed that 12-month poor quality of life was significantly associated with age > 65 years (OR 4.77), male gender (OR 0.44), pre-injury health problems (OR 2.30), admission to ICU (OR 2.15), ISS score 26–40 (OR 3.72), baseline PCS (OR 0.89), one-month PCS (OR 0.89), one-month MCS (OR 0.97), 6-month PCS (OR 0.76) and 6-month MCS (OR 0.97). Conclusion: For patients sustaining moderate or major trauma in Hong Kong at 12 months after injury < 1 in 10 patients had an excellent recovery, 3 in 10 reached a physical health status score Hong Kong norm, although as many as 6 in 10 patients had a mental health status score which is Hong Kong norm. ß 2013 Elsevier Ltd. All rights reserved.
Keywords: Functional outcome Morbidity Quality of life Trauma Wounds and injuries
Introduction
* Corresponding author at: Accident and Emergency Medicine Academic Unit, Chinese University of Hong Kong, Shatin, New Territories, Hong Kong. Tel.: +852 26321034; fax: +852 26481469. E-mail address: [email protected] (T.H. Rainer). 0020–1383/$ – see front matter ß 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.injury.2013.11.006
The current worldwide trend to implement trauma systems has aimed to improve survival, quality of life, and functional outcome in trauma patients, and there is good evidence that this has effectively reduced mortality after trauma [1–3]. As a result, the Hospital Authority in Hong Kong, in 2003, designated five hospitals as trauma centres [4], and these changes have resulted in improved
T.H. Rainer et al. / Injury, Int. J. Care Injured 45 (2014) 902–909
survival after trauma in Hong Kong [5,6]. However, these improvements have now begun to plateau. Patient-centred, health-related outcomes are increasingly recognised as a true benchmark of the quality of care received, and survivors of trauma patients often experience late sequelae that have a major impact on almost all aspects of their everyday life [7–9]. Quality of survival ranks as high priority for trauma system research [10]. Although limited to a very few centres, there is some data on post-trauma quality of survival [11–17], and some very limited evidence that the implementation of a trauma system improves post-trauma functional recovery and quality of life in patients with major lower limb trauma, and in major trauma in general [18,19]. Although trauma systems focus on major trauma, many patients in the system have moderate injury, which has not been evaluated. There is little data using the full short-form questionnaire evaluation, and no data to our knowledge on the Chinese, the most populous ethnic group on planet Earth. Further, one of the greatest concerns for trauma survivors and their relatives relate to prognosis [20]. What are the chances of a good outcome, and of a good quality of life? An evaluation of functional recovery and quantification of the burden of non-fatal trauma is important as this will allow comparison with other settings, will help evaluate the impact and effectiveness of the trauma system as a whole, and may provide some prognostic information for healthcare workers and patients. Currently there is published data on the health status of the normal population in Hong Kong [21,22] but there is little or no data for post-trauma health status and function for moderate and severe trauma in Chinese in general, and in Hong Kong in particular. It is important to evaluate the current quality of life in trauma survivors as the new trauma system takes shape. The aim of this prospective multicentre, cohort study in moderate and major trauma patients in Hong Kong was firstly to evaluate physical and mental post-trauma health status during the first year after injury, secondly to evaluate overall functional outcome, and thirdly to identify predictors of 12-month quality of survival. Materials and methods Study design This is a multi-centre prospective cohort study of trauma patients admitted to the Prince of Wales Hospital (PWH), Queen Elizabeth Hospital (QEH) and Tuen Mun Hospital (TMH) in Hong Kong. Ethical approval was obtained from the local ethics review board. Setting Trauma patients admitted to three regional hospitals and designated trauma centres located in the Eastern New Territories, the Western New Territories and the Kowloon region of Hong Kong – the Prince of Wales Hospital (PWH), Queen Elizabeth Hospital (QEH) and Tuen Mun Hospital (TMH) – were included. In 2007 the Emergency Departments of these hospitals received 556, 893 patients of whom 87, 543 (16%) were trauma patients. All trauma patients are included in the trauma registry if firstly they were triaged as either critical (category 1) or emergency (category 2) and were therefore managed in the emergency department trauma resuscitation rooms; secondly if they were admitted to the intensive care unit (ICU); or thirdly if they died after admission. In 2010, 1539 trauma patients were included in the registry and 505 (32.8%) patients had an Injury Severity Score 16. The trauma registries of these three hospitals were used to dichotomise trauma patients into major (ISS 16) and moderate (ISS 9–14) injury.
903
Generalisability A Hong Kong wide trauma database was set up in 2004 with data collected from all five trauma centres. The M statistic is a term which refers to the relative proportion of injury severity groups in a database or population. An M statistic of between 0.88 and 1.0 implies a similar mix of injury type and severity, and the reference is usually the United States spectrum of trauma. The M statistic [Hospital Authority Annual Trauma Report, unpublished data] for trauma as a whole in Hong Kong in 2005 was 0.972. The M statistics for each of the three trauma centres included in this study in 2010 were PWH 0.92, QEH 0.90 and TMH 0.94. This means that all three centres have a similar mix of trauma, which is also similar to the US as a whole. Patients, inclusion and exclusion criteria Adult patients aged 18 years with moderate or major trauma (defined as an ISS 9) who are entered into the trauma registry were included. The following patients were excluded: patients who were likely to leave Hong Kong; patients who are unwilling to enter the study; patients with ISS < 9; patients with isolated hip or pathological fractures. A record was kept of all patients who were not enrolled and also of the reasons for exclusion. Measurements and data collection In this study, the physical and mental health status of trauma patients was evaluated using the physical component summary (PCS) score and mental component summary (MCS) scores, respectively, of the generic Short-Form 36 (SF36) version 1 health status tool [23,24] and extended Glasgow Outcome Scale (GOSE) [25,26]. The SF36 is well-validated, reliable and sensitive to change and has been extensively used to assess and follow up trauma patients [27,28]. There are UK, US, Australasian, Chinese and Hong Kong specific ‘population norms’ for the major subdivisions and subscales of the SF36 which allow meaningful comparisons of health status between the group of interest and the general population [21,22]. The HK norm for the PCS is 52.83, and for the MCS is 47.18. This compares with the US norm, which is 50 for both PCS and MCS. Demographic data including age, sex, and mechanism of injury, Injury Severity Score (ISS) [29,30], Revised Trauma Score (RTS) [31], probability of survival (Ps) [32], Glasgow Coma Scale (GCS), hospital and ICU length of stay (LOS) and contact information were extracted from the trauma registry and patients’ records. Each trauma centre has a fully trained and dedicated trauma nurse coordinator who has been collecting injury data for the last five years. All of our trauma nurse coordinators were trained in AIS scoring in the USA and therefore ISS, RTS and Ps were calculated according to standard techniques. Injured patients were classified according to whether injuries were isolated or multiple, and according to specific body regions – head and neck injury, chest injury, abdominal injury and extremity injury. Isolated injury was defined as a single AIS 3. Multiple injury was defined as two or more regions with AIS 3. Working status was defined as a GOSE 5. Four periods were selected for assessment: (a) a ‘baseline’ assessment; (b) discharge/30 day assessment; (c) 6 month assessment; and (d) a 12 month assessment. Patients were recruited and followed up by trained research nurses. At each of the following assessment points, SF36 and GOSE questionnaires were completed. Although every effort was made to collect data directly from each patient, some patients were unable to give a meaningful reply. Under such circumstances, information from the closest relative was used.
904
T.H. Rainer et al. / Injury, Int. J. Care Injured 45 (2014) 902–909
For the baseline and pre-injury assessment, patients were contacted within one week of injury, usually during their inpatient stay, in order to obtain consent and initial data (baseline status). The SF-36 questionnaire asks some questions relating to the immediate time of assessment and others that refer to a patient’s best response in the previous four weeks. Subsequent follow up assessments involved two strategies. Firstly, the research nurse made up to five attempts to contact the patient by telephone and to conduct a questionnaire interview. Secondly, as Hong Kong has a well-integrated, centralised, computerised medical data system, it is possible to locate patient data on line, including whether they are alive or dead, discharged and/or readmitted, attending outpatient clinics, and if so, their health status at the time. The final outcome measurement in this study was determined at 12 months post injury. Outcomes and endpoints The primary outcome is post-traumatic SF36 score (PCS and MCS) assessed at 12 months after injury. The secondary outcome is GOSE also assessed at 12 months after injury. Poor health status was defined as a PCS of 25 or lower at 12 months post-injury. Statistical analysis Parametric and non-parametric descriptive statistics were used to summarise data depending on whether data are normally distributed or not. Chi-square and Fisher’s exact tests were used for categorical data whilst the t-test was used to compare means of continuous variables. A value of p < 0.05 is considered to be statistically significant and all tests were two tailed. A univariate analysis was conducted on the data set at each time point. Raw data from each SF36 questionnaire were transformed using standard evaluation techniques available from the developers of the scores [33] and recoded into the eight health domains and the MCS and PCS summary scores. This gives rise to continuous variables for the summary scores and for the eight health domains, which can be compared using standard one and two sample t-tests. Data from GOSE are categorical and can be compared using the Chi square or Fisher’s exact tests. Repeated measures analyses were then used to examine trends in each of the summary scores, domains of the SF36, and changes in GOSE. Further analyses focussed on specific groups. Attention was paid to anatomical injury including isolated versus multiple injuries, degree of injury, and specific body regions (head, spine, chest, abdomen, limbs). Results Between 1 January 2010 and 31 September 2010, 400 patients (mean age 53.3 years; range 18–106; 69.5% male) were recruited to the study. Table 1 shows the baseline characteristics of the 400 trauma patients from the three trauma centres. The commonest mechanisms of injury were falls and motor vehicle crashes. Over 65% of recruited patients had major trauma (i.e. an ISS > 15). Ninety-four percent of cases had a probability of survival >50%. For readers who are interested in greater detail, supplementary data files show the abbreviated injury scores for each body region (Table A1), and successful follow up rates at 1, 6 and 12 months after injury (Table A2). GOSE scores were available at 1, 6 and 12 months after injury for 85%, 70% and 59% cases, respectively, whilst SF36 scores were available for 75%, 56% and 44% cases, respectively. Table 2 compares the baseline characteristics between responders, non-responder survivors, and patients who died. There
Table 1 Baseline characteristics (N = 400).
Age Mean (SD) Age 18–35 Age 36–45 Age 46–55 Age 56–65 Age >65 Sex Female Male Pre-existing morbidity Good past health With existing health problems Missing/unknown Mechanism of injury Fall Fall (40
TMH subjects (N = 57)
74 63 26 8
106 (61.6%) 66 (38.4%)
28 (49.1%) 29 (50.9%)
(43.3%) (36.8%) (15.2%) (4.7%)
40 83 43 6
(23.3%) (48.3%) (25.0%) (3.5%)
25 15 14 3
(43.9%) (26.3%) (24.6%) (5.3%)
Probability of survival (Ps) Ps 0.96–1.00 111 (64.9%) Ps 0.91–0.95 19 (11.1%) Ps 0.76–0.90 23 (13.5%) Ps 0.51–0.75 3 (1.8%) Ps 0.26–0.50 3 (1.8%) Ps 0–0.25 3 (1.8%)
77 43 34 6 10 2
(44.8%) (25.0%) (19.8%) (3.5%) (5.8%) (1.1%)
34 9 7 3 2 2
(59.6%) (15.8%) (12.3%) (5.3%) (3.5%) (3.5%)
were no statistically significant differences between responders and non-responders. However, there were significant differences between these groups and the group of patients who died. Table 3 shows the GOSE levels of the 400 patients by longitudinal follow up to 1 year. At 12 months post-injury, 237/ 400 cases were available for analysis. Only 39/237 (16.5%) cases reported excellent recovery (GOSE = 8) whilst 76/237 (32.1%) cases had reached a GOSE 7. The relationship between baseline GOSE and 12 month mortality is shown in Table 4. Table 4 shows the changes in GOSE scores between baseline and at 12 month follow up. The drop out (or failure to follow up) rate was 45% in the groups with GOSE of 3 or 4, and 28% in patients with baseline GOSE = 2. It is unlikely that these represent deaths as Hong Kong has a comprehensive centralised medical database for its population. The majority of patients with baseline GOSE 7, who were followed up at 12 months, maintained their GOSE level although 9.5% showed a decrease. In those patients with a baseline GOSE of 5 or 6, 16/23 (70%) maintained or improved their level of GOSE. 37/70 (52.9%) deaths were in cases who had a baseline GOSE = 2. Of the cases who died, 37/60 (62%) had a baseline
T.H. Rainer et al. / Injury, Int. J. Care Injured 45 (2014) 902–909
905
Table 2 Characteristics of trauma patients presented as patients completing follow up at 12 months, non-responder survivors, and patients who died (N = 400). Responders who survived* (N = 177) Age Mean age (SD)
Non-responders who survived (N = 163)
52.9 (19.9)
49.2 (19.2)
Sex Female Male
52 (29.4%) 125 (70.6%)
52 (32.0%) 111 (68.0%)
Pre-existing morbidity Good past health With existing health problems
110 (62.3%) 65 (36.6%)
106 (65.0%) 55 (33.8%)
ISS Moderate (ISS 9–14) Major (ISS 16)
68 (38.4%) 109 (61.6%)
64 (39.3%) 99 (60.7%)
73 (41.3%) 25 (14.1%) 11 (6.2%)
64 (39.2%) 27 (16.6%) 8 (4.9%)
65 (36.7%) 24 (13.6%)
50 (30.8%) 21 (12.9%)
8 9 23 11 7 6 4 20
8 11 22 11 9 15 5 11
Other ISS groups ISS 16–25 ISS 26–35 ISS > 35 Mechanism of injury Fall ( 47.18 (HK norm) N < 47.18 (HK norm) Spine Number available for analysis PCS N > 52.83 (HK norm) N < 52.83 (HK norm) MCS N > 47.18 (HK norm) N < 47.18 (HK norm) Extremity Number available for analysis PCS N > 52.83 (HK norm) N < 52.83 (HK norm) MCS N > 47.18 (HK norm) N < 47.18 (HK norm) Thorax Number available for analysis PCS N > 52.83 (HK norm) N < 52.83 (HK norm) MCS N > 47.18 (HK norm) N < 47.18 (HK norm) Abdomen Number available for analysis PCS N > 52.83 (HK norm) N < 52.83 (HK norm) MCS N > 47.18 (HK norm) N < 47.18 (HK norm) Others Number available for analysis PCS N > 52.83 (HK norm) N < 52.83 (HK norm) MCS N > 47.18 (HK norm) N < 47.18 (HK norm)
Baseline
1 month
6 month
12 month
400 0 (0.0%) 0 (0.0%) 400
400 65 (16.3%) 37(9.2%) 298
400 122 (30.5) 55 (13.8%) 223
400 164 (41.0%) 60 (15.0%) 176
400 (100.0%)
298 (74.5%)
223 (55.7%)
176 (44.0%)
19 (4.8%) 374 (93.5%)
27 (6.7%) 271 (67.8%)
60 (15.0%) 163 (40.7%)
62 (15.5%) 114 (28.5%)
228 (57.0%) 165 (41.3%)
114 (28.5%) 184 (46.0%)
159 (39.7%) 64 (16.0%)
125 (31.2%) 51 (12.8%)
230 (100.0%)
162 (70.4%)
121 (52.6%)
100 (43.5%)
10 (4.3%) 220 (95.7%)
19 (8.2%) 143 (62.2%)
38 (16.5%) 83 (36.1%)
41 (17.8%) 59 (25.7%)
116 (50.4%) 114 (49.6%)
57 (24.8%) 105 (45.6%)
92 (40.0%) 29 (12.6%)
71 (30.9%) 29 (12.6%)
8 (100.0%)
8 (100%)
4 (50.0%)
3 (37.5%)
0 (0.0%) 8 (100.0%)
0 (0.0%) 8 (100.0%)
0 (0.0%) 4 (50.0%)
0 (0.0%) 3 (37.5%)
7 (87.5%) 1 (12.5%)
2 (25.0%) 6 (75.0%)
3 (37.5%) 1 (12.5%)
1 (12.5%) 2 (25.0%)
83 (100.0%)
67 (80.7%)
51 (61.4%)
41 (49.4%)
3 (3.6%) 80 (96.4%)
1 (1.2%) 66 (79.5%)
9 (10.8%) 42 (50.6%)
9 (10.8%) 32 (38.6%)
61 (73.5%) 22 (26.5%)
34 (41.0%) 33 (39.7%)
37 (44.6%) 14 (16.8%)
31 (37.3%) 10 (12.1%)
49 (100.0%)
41 (83.7%)
31 (63.3%)
23 (46.9%)
4 (8.2%) 45 (91.8%)
1 (2.1%) 40 (81.6%)
7 (14.3%) 24 (49.0%)
8 (16.3%) 15 (30.6%)
31 (63.3%) 18 (36.7%)
9 (18.4%) 32 (65.3%)
16 (32.7%) 15 (30.6%)
16 (32.6%) 7 (14.3%)
11 (100.0%)
10 (90.9%)
8 (72.8%)
5 (45.5%)
2 (18.2%) 9 (81.8%)
3 (27.3%) 7 (63.6%)
4 (36.4%) 4 (36.4%)
2 (18.2%) 3 (27.3%)
6 (54.5%) 5 (45.5%)
7 (63.6%) 3 (27.3%)
6 (54.6%) 2 (18.2%)
3 (27.3%) 2 (18.2%)
12 (100.0%)
10 (83.3%)
8 (66.7%)
4 (33.4%)
0 (0.0%) 12 (100.0%)
3 (25.0%) 7 (58.3%)
2 (16.7%) 6 (50.0%)
2 (16.7%) 2 (16.7%)
7 (58.3%) 5 (41.7%)
3 (25.0%) 7 (58.3%)
5 (41.7%) 3 (25.0%)
3 (25.0%) 1 (8.4%)
Note that all percentages relate to the maximum available at baseline, whether for the whole cohort or for each body region.
A univariate analysis showed that long-term (12-month) poor quality of life was significantly associated with age greater than 65 years (OR 4.77, 95% CI 1.61–14.20, p = 0.005), male gender (OR 0.44, 95% CI 0.22–0.90, p = 0.024), pre-injury health problems (OR 2.30, 95% CI 1.14–4.61, p = 0.019), admission to ICU (OR 2.15, 95% CI 1.07–4.34, p = 0.032), ISS
score 26–40 (OR 3.72, 95% CI 1.42–9.75, p = 0.007), baseline PCS (OR 0.89, 95% CI 0.85–0.94, p < 0.0001), one-month PCS (OR 0.89, 95% CI 0.85–0.93, p < 0.0001), one-month MCS (OR 0.97, 95% CI 0.94–1.00, p = 0.043), 6-month PCS (OR 0.76, 95% CI 0.69– 0.83, p < 0.0001) and 6-month MCS (OR 0.97, 95% CI 0.94–1.00, p = 0.027).
908
T.H. Rainer et al. / Injury, Int. J. Care Injured 45 (2014) 902–909
Discussion This is the first study to our knowledge to track baseline and functional recovery up to one year after injury for moderately and severely injured Chinese patients, and to use the full SF-36 quality of life questionnaire for assessment. It responds to the call that ‘efforts are needed to quantify the population burden of nonfatal injury and further our knowledge of the impact of trauma systems and trauma centre care on the quality of survival of trauma patients’ [19]. It provides important information on the effectiveness of the trauma system from three designated trauma centres from Hong Kong, giving patient-centred, health-related outcomes that allow comparison with trauma services globally. The quality of survival outcomes utilise both SF36 and GOSE assessments, which are rarely reported for a single trauma population at one year after injury. The case-mix of trauma in Hong Kong is similar to that of the US in general as evidenced by M statistics of 0.90 or greater, and blunt mechanisms of injury characterised predominantly by motor vehicle crashes and falls. However, the burden and volume of trauma presenting to trauma centres in Hong Kong is much less than that of a level 1 trauma centre in the US. Post-trauma physical function improves gradually for up to 12 months after injury but less than half of all recruited cases reached the level of the PCS for the Hong Kong population norm. This holds true for both isolated and multiple trauma, and also for each body region studied. At 12 months, only 19% cases had reached a high functional outcome (i.e. GOSE 7), although a further 11% had upper moderate disability (i.e. GOSE = 6). This compares poorly with data from Utrecht in the Netherlands where 79.9% cases actual returned either to part or to full-time work [34] and to Victoria in Australia where 57% returned either to part or to full-time work. Improvements in the MCS are such that in over half of cases they exceed that of the Hong Kong population norm for all injury groups by 6 months. It is possible that the awareness that they have survived significant trauma has a very positive effect, leaving patients with a sense of mental well-being that is greater than the population as a whole. The one group – spinal trauma – with a lower MCS at one year may be due to an awareness of prolonged and possibly non-recoverable injury. After univariate analysis, the significant predictors of long-term (12-month), poor quality of life were age greater than 65 years, male gender, pre-injury health problems, admission to ICU, ISS, baseline PCS, one-month PCS and MCS, and 6-month PCS and MCS. The successful follow up rate at 12 months in Hong Kong was 59%. At least five attempts were made to contact patients by telephone but without success. However, Hong Kong as an integrated medical system whereby it is easy to determine whether the patient is still alive or not. Although it was not possible to evaluate health status in 41% cases, nevertheless the patient was still alive. The proportion of patients with upper good recovery and upper moderate disability at one year after injury is less than those reported from Victoria, Australia [19]. Although we found similar predictors of poorer functional outcome as other studies, such as higher age and comorbidities, nevertheless there were significant differences. For example, female gender did not predict a worse outcome. These differences may result partly from the fact that the Australian population includes only major injury, whilst the Hong Kong population includes both moderate and major injury. Further direct comparison between such centres allowing for important variable interaction would facilitate a useful evaluation of the effectiveness of trauma care. Although the sample size is relatively small, and collected by convenience rather than consecutively, nevertheless this is a landmark study for quality of survival in Asia and for Chinese
patients. The successful follow up rates are less than those reported in more recently published studies [19] but are still of a high level. We did not study the effect of socioeconomic status, educational level or compensation all of which may affect of recovery. The low follow up rate and the reasons for subjects not responding is always a concern in such studies. People may not respond for many extremely different reasons. These include death on the one hand, and excellent recovery on the other. Those with excellent recovery may drop out because they do not wish to go through the trouble of further assessment. Thus there is a risk that the analysis has an inherent systematic bias that among those who did not complete follow up evaluation, there was a higher percentage of excellent recovery. This could be explained by the well recovered patient has less incentive to maintain contact with health care providers. How are we to sort out the impact of these two opposite but potential systematic biases? As already mentioned, Hong Kong has a robust, centralised healthcare database of all patients attending public hospitals. From this database it is possible to follow the survival, reattendance, readmission and clinic records of all cases. Although data from these records are not designed for our study, nevertheless much can be gleaned about the crude survival and functional state of the patients. Data in Table 2 shows the proportion of non-responders at 12 months after injury that were still alive, and therefore at least death is not the reason in the 163 (73%) surviving non-responders. One of the challenges in a study of this nature, is that readers and stakeholders may have different perspectives, and the answers to their questions are affected by the proportion of missing data. Some readers may want to know the change in recovery from baseline, whilst others are interested in specific regional injuries. Thus it is not always clear whether data should be best analysed with reference to the baseline set, or to the available follow up set. The issue could be clearer if each separate analysis was presented as a separate table, but this would increase the number of tables to 12 or 14, which would be unacceptable to the journal. Where possible we have tried to clarify the issue by clearly presenting raw data as numerators and denominators, and giving a clear indication of the primary reference which is stated as ‘100%’. This reference may change depending on whether the reference is the original baseline data, the available data at 1, 6 or 12 month follow up, and whether the analysis is for the dataset for all injuries or for focussed regional injury. As we have presented raw data, then readers with a different perspectives may perform alternative analyses if they wish to obtain percentages that answer their particular question. After the primary analysis of the data, a sensitivity analysis was considered. However, as 12-month responders and surviving nonresponders have similar baseline characteristics, it is reasonable to assume that the proportion of surviving non-responders exceeding the HK norm will be similar to the proportion of responders exceeding the HK norm. The reasons why people may not respond are manifold, but probably includes some who are too well and cannot be bothered to respond, and those patients who are too ill and unable to respond. As contact was not possible, it is not possible to say which if any group predominates. What can reasonably be determined is that death is an unlikely reason for no response, and the baseline characteristics in terms of age, gender, pre-existing morbidity, anatomical and physiological derangement between responders and non-responders are very similar. With this in mind, the conclusions of this study reflect our view that as surviving non-responders have similar baseline characteristics to surviving responders, and the proportion of surviving nonresponders to surviving responders is almost 1:1, then the final recovery of the whole population studied may be double that presented for surviving responders alone.
T.H. Rainer et al. / Injury, Int. J. Care Injured 45 (2014) 902–909
It is important to note that improvements in quality of survival do not appear to have reached a plateau at one year after injury and further study is required for longer term effects. Conclusions For patients sustaining moderate or major trauma in Hong Kong at 12 months after injury less than 1 in 10 patients had an excellent recovery, less than or equal to 3 in 10 patients reached a physical health status score greater or equal to the Hong Kong population norm, although as many as 6 in 10 patients had a mental health status score which is greater or equal to the Hong Kong norm. Conflict of interest statement None. Acknowledgments We thank the Health and Health Services Research Grant Committee of Hong Kong Government for their generous support of this study (HHSRF 07080261). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.injury.2013.11.006. References [1] Sethi D, Aljunid S, Saperi SB, Clemens F, Hardy P, Elbourne D, et al. Comparison of the effectiveness of trauma services provided by secondary and tertiary hospitals in Malaysia. Ann Emerg Med 2007;49:52–61. [2] Tinkoff GH, O’Connor RE, Alexander EL, Jones MS. The Delaware trauma system: impact of level III trauma centers. J Trauma 2007;63:121–7. [3] Rainer TH, Smit P, de V. Trauma systems and emergency medicine. Emerg Med 2003;15:11–7. [4] Cheung NK, Yeung JHH, Chan JTS, Cameron PA, Graham CA, Rainer TH. Primary trauma diversion: initial experience in Hong Kong. J Trauma 2006;61:954–60. [5] Leung GK, Chang A, Cheung FC, Ho HF, Ho W, Hui SM, et al. The first 5 years since trauma center designation in the Hong Kong Special Administrative Region, People’s Republic of China. J Trauma 2011;70:1128–33. [6] Cheng CH, Graham CA, Gabbe BJ, Yeung JH, Kossmann T, Judson RT, et al. Trauma care systems: a comparison of trauma care in Victoria, Australia and Hong Kong, China. Ann Surg 2008;247(2):335–42. [7] Moi AL, Wentzel-Larsen T, Salemark L, Wahl AK, Hanestad BR. Impaired generic health status but perception of good quality of life in survivors of burn injury. J Trauma 2006;61:961–8. [8] Kiely JM, Brasel KJ, Weidner KL, Guse CE, Weigelt JA. Predicting quality of life six months after traumatic injury. J Trauma 2006;61:791–8. [9] Graham CA, Gordon MWG, Roy CW, Hanion PW. Long term outcomes of major trauma without head injury in the West of Scotland: pilot case control study. Eur J Emerg Med 2007;14:35–8. [10] Mann N, Mullins R. Research recommendations and proposed action items to facilitate trauma system implementation and evaluation. J Trauma 1999;47: S75–8.
909
[11] Utomo WK, Gabbe BJ, Simpson PM, Cameron PA. Predictors of in-hospital mortality and 6-month functional outcomes in older adults after moderate to severe traumatic brain injury. Injury 2009;40:973–7. [12] Gabbe BJ, Sutherland AM, Williamson OD, Cameron PA. Use of health care services 6 months following major trauma. Aust Health Rev 2007;31:628–32. [13] O’Mullane PA, Mikocka-Walus AA, Gabbe BJ, Cameron PA. Incidence and outcomes of major trauma assaults: a population-based study in Victoria. Med J Aust 2009;190:129–32. [14] Gruen RL, Gabbe BJ, Stelfox HT, Cameron PA. Indicators of the quality of trauma care and the performance of trauma systems. Br J Surg 2012;99(Suppl. 1):97– 104. http://dx.doi.org/10.1002/bjs.7754. [15] Gabbe BJ, Simpson PM, Sutherland AM, Palmer CS, Williamson OD, Butt W, et al. Functional and health-related quality of life outcomes after pediatric trauma. J Trauma 2011;70:1532–8. [16] Williamson OD, Gabbe BJ, Sutherland AM, Wolfe R, Forbes AB, Cameron PA. Comparing the responsiveness of functional outcome assessment measures for trauma registries. J Trauma 2011;71:63–8. [17] Nichol AD, Higgins AM, Gabbe BJ, Murray LJ, Cooper DJ, Cameron PA. Measuring functional and quality of life outcomes following major head injury: common scales and checklists. Injury 2011;42:281–7. [18] MacKenzie EJ, Rivara FP, Jurkovitch GJ, Nathens AB, Egleston BL, Slaveker DS, et al. The impact of trauma-center care on functional outcomes following major lower limb trauma. J Bone Joint Surg 2008;90-A:101–9. [19] Gabbe BJ, Simpson PM, Sutherland AM, Wolfe R, Fitzgerald MC, Judson R, et al. Improved functional outcomes for major trauma patients in a regionalized, inclusive trauma system. Ann Surg 2012;255:1009–15. [20] Gabbe BJ, Sleney J, Gosling C, Wilson K, Sutherland A, Hart M, et al. Exploring trauma patient perception of barriers and facilitators to recovery. In: Conference Proceedings of the Annual Scientific Meeting of the Australasian Trauma Society. Trauma 2012: Resuscitation to Rehabilitation; 2012. p. A35. [21] Lam CL, Lauder IJ, Lam TP, Gandek B. Population based norming of the Chinese (HK) version of the SF-36 health survey. Hong Kong Pract 1999;21: 460–70. [22] Lam CL, Tse EYY, Gandek B, Fong DY. The SF-36 summary scales were valid, reliable, and equivalent in a Chinese population. J Clin Epidemiol 2005;58:815–22. [23] Bowling A. Measuring health: a review of quality of life measurement scales. Buckingham: Open University Press; 1997. [24] Garratt A, Schmidt L, Mackintosh A, Fitzpatrick R. Quality of life measurement: bibliographic study of patient assessed health outcome measures. Br Med J 2002;324:1417. [25] Nijboer JMM, van der Sluis CK, van der Naalt J, Nijsten MW, Ten Duis HJ. Two cohorts of severely injured trauma patients, nearly two decades apart: unchanged mortality but improved quality of life despite higher age. J Trauma 2007;63:670–5. [26] Wilson JTL, Pettigrew LEL, Teasdale GM. Structured interviews for the Glasgow Outcome Scale and Extended Glasgow Outcome Scale: guidelines for their use. J Neurotrauma 1998;15:573–85. [27] Brenneman FD, Redelmeier DA, Boulanger BR, McLellan BA, Culhane JP. Longterm outcomes in blunt trauma: who back to work? J Trauma 1997;42: 778–81. [28] MacKenzie EJ, McCarthy ML, Ditunno JF, Forrester-Staz C, Gruen GS, Marion DW, et al. Using the SF36 for characterising outcome after multiple trauma involving head injury. J Trauma 2002;52:527–34. [29] Baker SP, O’Neill B, Haddon Jr W, Long WB. The Injury Severity Score: a method for describing patients with multiple injuries and evaluating emergency care. J Trauma 1974;14:187–96. [30] Association for the advancement of automative medicine. In: The abbreviated injury scale. Des Plaines Ilinois; 1990 Revision. [31] Champion HR, Sacco WJ, Copes WS, Gann DS, Gennarelli TA, Flanagan ME. A revision of the Trauma Score. J Trauma 1989;29(5):623–9. [32] Boyd CR, Tolson MA, Copes WS. Evaluating trauma care: the TRISS method. Trauma Score and the Injury Severity Score. J Trauma 1987;27:370–8. [33] http://www.qualitymetric.com/ [accessed 10.08.12]. [34] Holtslag HR, Post MW, van der Werken C, Lindeman E. Return to work after major trauma. Clin Rehabil 2007;21:373–83.
Contents lists available at ScienceDirect
Injury journal homepage: www.elsevier.com/locate/injury
Assessment of quality of life and functional outcome in patients sustaining moderate and major trauma: A multicentre, prospective cohort study T.H. Rainer a,b,*, J.H.H. Yeung a,b, S.K.C. Cheung a, Y.K.Y. Yuen a, W.S. Poon c, H.F. Ho d, C.W. Kam e, G.N. Cattermole f, A. Chang d, F.L. So e, C.A. Graham a,b a
Accident and Emergency Medicine Academic Unit, Chinese University of Hong Kong, Hong Kong Trauma & Emergency Centre, Prince of Wales Hospital, Hong Kong c Division of Neurosurgery, Department of Surgery, Chinese University of Hong Kong, Hong Kong d Accident and Emergency Department, Queen Elizabeth Hospital, Hong Kong e Accident and Emergency Department, Tuen Mun Hospital, Hong Kong f Emergency Department, Princess Royal University Hospital, Orpington, UK b
A R T I C L E I N F O
A B S T R A C T
Article history: Accepted 10 November 2013
Background: Trauma care systems aim to reduce both death and disability, yet there is little data on posttrauma health status and functional outcome. Objectives: To evaluate baseline, discharge, six month and 12 month post-trauma quality of life, functional outcome and predictors of quality of life in Hong Kong. Methods: Multicentre, prospective cohort study using data from the trauma registries of three regional trauma centres in Hong Kong. Trauma patients with an ISS 9 and aged 18 years were included. The main outcome measures were the physical component summary (PCS) score and mental component summary (MCS) scores of the Short-Form 36 (SF36) for health status, and the extended Glasgow Outcome Scale (GOSE) for functional outcome. Results: Between 1 January 2010 and 31 September 2010, 400 patients (mean age 53.3 years; range 18– 106; 69.5% male) were recruited to the study. There were no statistically significant differences in baseline characteristics between responders (N = 177) and surviving non-responders (N = 163). However, there were significant differences between these groups and the group of patients who died (N = 60). Only 16/400 (4%) cases reported a GOSE 7. 62/400 (15.5%) responders reached the HK population norm for PCS. 125/400 (31%) responders reached the HK population norm for MCS. If nonresponders had similar outcomes to responders, then the percentages for GOSE 7 would rise from 4% to 8%, for PCS from 15.5% to 30%, and for MCS from 31% to 60%. Univariate analysis showed that 12-month poor quality of life was significantly associated with age > 65 years (OR 4.77), male gender (OR 0.44), pre-injury health problems (OR 2.30), admission to ICU (OR 2.15), ISS score 26–40 (OR 3.72), baseline PCS (OR 0.89), one-month PCS (OR 0.89), one-month MCS (OR 0.97), 6-month PCS (OR 0.76) and 6-month MCS (OR 0.97). Conclusion: For patients sustaining moderate or major trauma in Hong Kong at 12 months after injury < 1 in 10 patients had an excellent recovery, 3 in 10 reached a physical health status score Hong Kong norm, although as many as 6 in 10 patients had a mental health status score which is Hong Kong norm. ß 2013 Elsevier Ltd. All rights reserved.
Keywords: Functional outcome Morbidity Quality of life Trauma Wounds and injuries
Introduction
* Corresponding author at: Accident and Emergency Medicine Academic Unit, Chinese University of Hong Kong, Shatin, New Territories, Hong Kong. Tel.: +852 26321034; fax: +852 26481469. E-mail address: [email protected] (T.H. Rainer). 0020–1383/$ – see front matter ß 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.injury.2013.11.006
The current worldwide trend to implement trauma systems has aimed to improve survival, quality of life, and functional outcome in trauma patients, and there is good evidence that this has effectively reduced mortality after trauma [1–3]. As a result, the Hospital Authority in Hong Kong, in 2003, designated five hospitals as trauma centres [4], and these changes have resulted in improved
T.H. Rainer et al. / Injury, Int. J. Care Injured 45 (2014) 902–909
survival after trauma in Hong Kong [5,6]. However, these improvements have now begun to plateau. Patient-centred, health-related outcomes are increasingly recognised as a true benchmark of the quality of care received, and survivors of trauma patients often experience late sequelae that have a major impact on almost all aspects of their everyday life [7–9]. Quality of survival ranks as high priority for trauma system research [10]. Although limited to a very few centres, there is some data on post-trauma quality of survival [11–17], and some very limited evidence that the implementation of a trauma system improves post-trauma functional recovery and quality of life in patients with major lower limb trauma, and in major trauma in general [18,19]. Although trauma systems focus on major trauma, many patients in the system have moderate injury, which has not been evaluated. There is little data using the full short-form questionnaire evaluation, and no data to our knowledge on the Chinese, the most populous ethnic group on planet Earth. Further, one of the greatest concerns for trauma survivors and their relatives relate to prognosis [20]. What are the chances of a good outcome, and of a good quality of life? An evaluation of functional recovery and quantification of the burden of non-fatal trauma is important as this will allow comparison with other settings, will help evaluate the impact and effectiveness of the trauma system as a whole, and may provide some prognostic information for healthcare workers and patients. Currently there is published data on the health status of the normal population in Hong Kong [21,22] but there is little or no data for post-trauma health status and function for moderate and severe trauma in Chinese in general, and in Hong Kong in particular. It is important to evaluate the current quality of life in trauma survivors as the new trauma system takes shape. The aim of this prospective multicentre, cohort study in moderate and major trauma patients in Hong Kong was firstly to evaluate physical and mental post-trauma health status during the first year after injury, secondly to evaluate overall functional outcome, and thirdly to identify predictors of 12-month quality of survival. Materials and methods Study design This is a multi-centre prospective cohort study of trauma patients admitted to the Prince of Wales Hospital (PWH), Queen Elizabeth Hospital (QEH) and Tuen Mun Hospital (TMH) in Hong Kong. Ethical approval was obtained from the local ethics review board. Setting Trauma patients admitted to three regional hospitals and designated trauma centres located in the Eastern New Territories, the Western New Territories and the Kowloon region of Hong Kong – the Prince of Wales Hospital (PWH), Queen Elizabeth Hospital (QEH) and Tuen Mun Hospital (TMH) – were included. In 2007 the Emergency Departments of these hospitals received 556, 893 patients of whom 87, 543 (16%) were trauma patients. All trauma patients are included in the trauma registry if firstly they were triaged as either critical (category 1) or emergency (category 2) and were therefore managed in the emergency department trauma resuscitation rooms; secondly if they were admitted to the intensive care unit (ICU); or thirdly if they died after admission. In 2010, 1539 trauma patients were included in the registry and 505 (32.8%) patients had an Injury Severity Score 16. The trauma registries of these three hospitals were used to dichotomise trauma patients into major (ISS 16) and moderate (ISS 9–14) injury.
903
Generalisability A Hong Kong wide trauma database was set up in 2004 with data collected from all five trauma centres. The M statistic is a term which refers to the relative proportion of injury severity groups in a database or population. An M statistic of between 0.88 and 1.0 implies a similar mix of injury type and severity, and the reference is usually the United States spectrum of trauma. The M statistic [Hospital Authority Annual Trauma Report, unpublished data] for trauma as a whole in Hong Kong in 2005 was 0.972. The M statistics for each of the three trauma centres included in this study in 2010 were PWH 0.92, QEH 0.90 and TMH 0.94. This means that all three centres have a similar mix of trauma, which is also similar to the US as a whole. Patients, inclusion and exclusion criteria Adult patients aged 18 years with moderate or major trauma (defined as an ISS 9) who are entered into the trauma registry were included. The following patients were excluded: patients who were likely to leave Hong Kong; patients who are unwilling to enter the study; patients with ISS < 9; patients with isolated hip or pathological fractures. A record was kept of all patients who were not enrolled and also of the reasons for exclusion. Measurements and data collection In this study, the physical and mental health status of trauma patients was evaluated using the physical component summary (PCS) score and mental component summary (MCS) scores, respectively, of the generic Short-Form 36 (SF36) version 1 health status tool [23,24] and extended Glasgow Outcome Scale (GOSE) [25,26]. The SF36 is well-validated, reliable and sensitive to change and has been extensively used to assess and follow up trauma patients [27,28]. There are UK, US, Australasian, Chinese and Hong Kong specific ‘population norms’ for the major subdivisions and subscales of the SF36 which allow meaningful comparisons of health status between the group of interest and the general population [21,22]. The HK norm for the PCS is 52.83, and for the MCS is 47.18. This compares with the US norm, which is 50 for both PCS and MCS. Demographic data including age, sex, and mechanism of injury, Injury Severity Score (ISS) [29,30], Revised Trauma Score (RTS) [31], probability of survival (Ps) [32], Glasgow Coma Scale (GCS), hospital and ICU length of stay (LOS) and contact information were extracted from the trauma registry and patients’ records. Each trauma centre has a fully trained and dedicated trauma nurse coordinator who has been collecting injury data for the last five years. All of our trauma nurse coordinators were trained in AIS scoring in the USA and therefore ISS, RTS and Ps were calculated according to standard techniques. Injured patients were classified according to whether injuries were isolated or multiple, and according to specific body regions – head and neck injury, chest injury, abdominal injury and extremity injury. Isolated injury was defined as a single AIS 3. Multiple injury was defined as two or more regions with AIS 3. Working status was defined as a GOSE 5. Four periods were selected for assessment: (a) a ‘baseline’ assessment; (b) discharge/30 day assessment; (c) 6 month assessment; and (d) a 12 month assessment. Patients were recruited and followed up by trained research nurses. At each of the following assessment points, SF36 and GOSE questionnaires were completed. Although every effort was made to collect data directly from each patient, some patients were unable to give a meaningful reply. Under such circumstances, information from the closest relative was used.
904
T.H. Rainer et al. / Injury, Int. J. Care Injured 45 (2014) 902–909
For the baseline and pre-injury assessment, patients were contacted within one week of injury, usually during their inpatient stay, in order to obtain consent and initial data (baseline status). The SF-36 questionnaire asks some questions relating to the immediate time of assessment and others that refer to a patient’s best response in the previous four weeks. Subsequent follow up assessments involved two strategies. Firstly, the research nurse made up to five attempts to contact the patient by telephone and to conduct a questionnaire interview. Secondly, as Hong Kong has a well-integrated, centralised, computerised medical data system, it is possible to locate patient data on line, including whether they are alive or dead, discharged and/or readmitted, attending outpatient clinics, and if so, their health status at the time. The final outcome measurement in this study was determined at 12 months post injury. Outcomes and endpoints The primary outcome is post-traumatic SF36 score (PCS and MCS) assessed at 12 months after injury. The secondary outcome is GOSE also assessed at 12 months after injury. Poor health status was defined as a PCS of 25 or lower at 12 months post-injury. Statistical analysis Parametric and non-parametric descriptive statistics were used to summarise data depending on whether data are normally distributed or not. Chi-square and Fisher’s exact tests were used for categorical data whilst the t-test was used to compare means of continuous variables. A value of p < 0.05 is considered to be statistically significant and all tests were two tailed. A univariate analysis was conducted on the data set at each time point. Raw data from each SF36 questionnaire were transformed using standard evaluation techniques available from the developers of the scores [33] and recoded into the eight health domains and the MCS and PCS summary scores. This gives rise to continuous variables for the summary scores and for the eight health domains, which can be compared using standard one and two sample t-tests. Data from GOSE are categorical and can be compared using the Chi square or Fisher’s exact tests. Repeated measures analyses were then used to examine trends in each of the summary scores, domains of the SF36, and changes in GOSE. Further analyses focussed on specific groups. Attention was paid to anatomical injury including isolated versus multiple injuries, degree of injury, and specific body regions (head, spine, chest, abdomen, limbs). Results Between 1 January 2010 and 31 September 2010, 400 patients (mean age 53.3 years; range 18–106; 69.5% male) were recruited to the study. Table 1 shows the baseline characteristics of the 400 trauma patients from the three trauma centres. The commonest mechanisms of injury were falls and motor vehicle crashes. Over 65% of recruited patients had major trauma (i.e. an ISS > 15). Ninety-four percent of cases had a probability of survival >50%. For readers who are interested in greater detail, supplementary data files show the abbreviated injury scores for each body region (Table A1), and successful follow up rates at 1, 6 and 12 months after injury (Table A2). GOSE scores were available at 1, 6 and 12 months after injury for 85%, 70% and 59% cases, respectively, whilst SF36 scores were available for 75%, 56% and 44% cases, respectively. Table 2 compares the baseline characteristics between responders, non-responder survivors, and patients who died. There
Table 1 Baseline characteristics (N = 400).
Age Mean (SD) Age 18–35 Age 36–45 Age 46–55 Age 56–65 Age >65 Sex Female Male Pre-existing morbidity Good past health With existing health problems Missing/unknown Mechanism of injury Fall Fall (40
TMH subjects (N = 57)
74 63 26 8
106 (61.6%) 66 (38.4%)
28 (49.1%) 29 (50.9%)
(43.3%) (36.8%) (15.2%) (4.7%)
40 83 43 6
(23.3%) (48.3%) (25.0%) (3.5%)
25 15 14 3
(43.9%) (26.3%) (24.6%) (5.3%)
Probability of survival (Ps) Ps 0.96–1.00 111 (64.9%) Ps 0.91–0.95 19 (11.1%) Ps 0.76–0.90 23 (13.5%) Ps 0.51–0.75 3 (1.8%) Ps 0.26–0.50 3 (1.8%) Ps 0–0.25 3 (1.8%)
77 43 34 6 10 2
(44.8%) (25.0%) (19.8%) (3.5%) (5.8%) (1.1%)
34 9 7 3 2 2
(59.6%) (15.8%) (12.3%) (5.3%) (3.5%) (3.5%)
were no statistically significant differences between responders and non-responders. However, there were significant differences between these groups and the group of patients who died. Table 3 shows the GOSE levels of the 400 patients by longitudinal follow up to 1 year. At 12 months post-injury, 237/ 400 cases were available for analysis. Only 39/237 (16.5%) cases reported excellent recovery (GOSE = 8) whilst 76/237 (32.1%) cases had reached a GOSE 7. The relationship between baseline GOSE and 12 month mortality is shown in Table 4. Table 4 shows the changes in GOSE scores between baseline and at 12 month follow up. The drop out (or failure to follow up) rate was 45% in the groups with GOSE of 3 or 4, and 28% in patients with baseline GOSE = 2. It is unlikely that these represent deaths as Hong Kong has a comprehensive centralised medical database for its population. The majority of patients with baseline GOSE 7, who were followed up at 12 months, maintained their GOSE level although 9.5% showed a decrease. In those patients with a baseline GOSE of 5 or 6, 16/23 (70%) maintained or improved their level of GOSE. 37/70 (52.9%) deaths were in cases who had a baseline GOSE = 2. Of the cases who died, 37/60 (62%) had a baseline
T.H. Rainer et al. / Injury, Int. J. Care Injured 45 (2014) 902–909
905
Table 2 Characteristics of trauma patients presented as patients completing follow up at 12 months, non-responder survivors, and patients who died (N = 400). Responders who survived* (N = 177) Age Mean age (SD)
Non-responders who survived (N = 163)
52.9 (19.9)
49.2 (19.2)
Sex Female Male
52 (29.4%) 125 (70.6%)
52 (32.0%) 111 (68.0%)
Pre-existing morbidity Good past health With existing health problems
110 (62.3%) 65 (36.6%)
106 (65.0%) 55 (33.8%)
ISS Moderate (ISS 9–14) Major (ISS 16)
68 (38.4%) 109 (61.6%)
64 (39.3%) 99 (60.7%)
73 (41.3%) 25 (14.1%) 11 (6.2%)
64 (39.2%) 27 (16.6%) 8 (4.9%)
65 (36.7%) 24 (13.6%)
50 (30.8%) 21 (12.9%)
8 9 23 11 7 6 4 20
8 11 22 11 9 15 5 11
Other ISS groups ISS 16–25 ISS 26–35 ISS > 35 Mechanism of injury Fall ( 47.18 (HK norm) N < 47.18 (HK norm) Spine Number available for analysis PCS N > 52.83 (HK norm) N < 52.83 (HK norm) MCS N > 47.18 (HK norm) N < 47.18 (HK norm) Extremity Number available for analysis PCS N > 52.83 (HK norm) N < 52.83 (HK norm) MCS N > 47.18 (HK norm) N < 47.18 (HK norm) Thorax Number available for analysis PCS N > 52.83 (HK norm) N < 52.83 (HK norm) MCS N > 47.18 (HK norm) N < 47.18 (HK norm) Abdomen Number available for analysis PCS N > 52.83 (HK norm) N < 52.83 (HK norm) MCS N > 47.18 (HK norm) N < 47.18 (HK norm) Others Number available for analysis PCS N > 52.83 (HK norm) N < 52.83 (HK norm) MCS N > 47.18 (HK norm) N < 47.18 (HK norm)
Baseline
1 month
6 month
12 month
400 0 (0.0%) 0 (0.0%) 400
400 65 (16.3%) 37(9.2%) 298
400 122 (30.5) 55 (13.8%) 223
400 164 (41.0%) 60 (15.0%) 176
400 (100.0%)
298 (74.5%)
223 (55.7%)
176 (44.0%)
19 (4.8%) 374 (93.5%)
27 (6.7%) 271 (67.8%)
60 (15.0%) 163 (40.7%)
62 (15.5%) 114 (28.5%)
228 (57.0%) 165 (41.3%)
114 (28.5%) 184 (46.0%)
159 (39.7%) 64 (16.0%)
125 (31.2%) 51 (12.8%)
230 (100.0%)
162 (70.4%)
121 (52.6%)
100 (43.5%)
10 (4.3%) 220 (95.7%)
19 (8.2%) 143 (62.2%)
38 (16.5%) 83 (36.1%)
41 (17.8%) 59 (25.7%)
116 (50.4%) 114 (49.6%)
57 (24.8%) 105 (45.6%)
92 (40.0%) 29 (12.6%)
71 (30.9%) 29 (12.6%)
8 (100.0%)
8 (100%)
4 (50.0%)
3 (37.5%)
0 (0.0%) 8 (100.0%)
0 (0.0%) 8 (100.0%)
0 (0.0%) 4 (50.0%)
0 (0.0%) 3 (37.5%)
7 (87.5%) 1 (12.5%)
2 (25.0%) 6 (75.0%)
3 (37.5%) 1 (12.5%)
1 (12.5%) 2 (25.0%)
83 (100.0%)
67 (80.7%)
51 (61.4%)
41 (49.4%)
3 (3.6%) 80 (96.4%)
1 (1.2%) 66 (79.5%)
9 (10.8%) 42 (50.6%)
9 (10.8%) 32 (38.6%)
61 (73.5%) 22 (26.5%)
34 (41.0%) 33 (39.7%)
37 (44.6%) 14 (16.8%)
31 (37.3%) 10 (12.1%)
49 (100.0%)
41 (83.7%)
31 (63.3%)
23 (46.9%)
4 (8.2%) 45 (91.8%)
1 (2.1%) 40 (81.6%)
7 (14.3%) 24 (49.0%)
8 (16.3%) 15 (30.6%)
31 (63.3%) 18 (36.7%)
9 (18.4%) 32 (65.3%)
16 (32.7%) 15 (30.6%)
16 (32.6%) 7 (14.3%)
11 (100.0%)
10 (90.9%)
8 (72.8%)
5 (45.5%)
2 (18.2%) 9 (81.8%)
3 (27.3%) 7 (63.6%)
4 (36.4%) 4 (36.4%)
2 (18.2%) 3 (27.3%)
6 (54.5%) 5 (45.5%)
7 (63.6%) 3 (27.3%)
6 (54.6%) 2 (18.2%)
3 (27.3%) 2 (18.2%)
12 (100.0%)
10 (83.3%)
8 (66.7%)
4 (33.4%)
0 (0.0%) 12 (100.0%)
3 (25.0%) 7 (58.3%)
2 (16.7%) 6 (50.0%)
2 (16.7%) 2 (16.7%)
7 (58.3%) 5 (41.7%)
3 (25.0%) 7 (58.3%)
5 (41.7%) 3 (25.0%)
3 (25.0%) 1 (8.4%)
Note that all percentages relate to the maximum available at baseline, whether for the whole cohort or for each body region.
A univariate analysis showed that long-term (12-month) poor quality of life was significantly associated with age greater than 65 years (OR 4.77, 95% CI 1.61–14.20, p = 0.005), male gender (OR 0.44, 95% CI 0.22–0.90, p = 0.024), pre-injury health problems (OR 2.30, 95% CI 1.14–4.61, p = 0.019), admission to ICU (OR 2.15, 95% CI 1.07–4.34, p = 0.032), ISS
score 26–40 (OR 3.72, 95% CI 1.42–9.75, p = 0.007), baseline PCS (OR 0.89, 95% CI 0.85–0.94, p < 0.0001), one-month PCS (OR 0.89, 95% CI 0.85–0.93, p < 0.0001), one-month MCS (OR 0.97, 95% CI 0.94–1.00, p = 0.043), 6-month PCS (OR 0.76, 95% CI 0.69– 0.83, p < 0.0001) and 6-month MCS (OR 0.97, 95% CI 0.94–1.00, p = 0.027).
908
T.H. Rainer et al. / Injury, Int. J. Care Injured 45 (2014) 902–909
Discussion This is the first study to our knowledge to track baseline and functional recovery up to one year after injury for moderately and severely injured Chinese patients, and to use the full SF-36 quality of life questionnaire for assessment. It responds to the call that ‘efforts are needed to quantify the population burden of nonfatal injury and further our knowledge of the impact of trauma systems and trauma centre care on the quality of survival of trauma patients’ [19]. It provides important information on the effectiveness of the trauma system from three designated trauma centres from Hong Kong, giving patient-centred, health-related outcomes that allow comparison with trauma services globally. The quality of survival outcomes utilise both SF36 and GOSE assessments, which are rarely reported for a single trauma population at one year after injury. The case-mix of trauma in Hong Kong is similar to that of the US in general as evidenced by M statistics of 0.90 or greater, and blunt mechanisms of injury characterised predominantly by motor vehicle crashes and falls. However, the burden and volume of trauma presenting to trauma centres in Hong Kong is much less than that of a level 1 trauma centre in the US. Post-trauma physical function improves gradually for up to 12 months after injury but less than half of all recruited cases reached the level of the PCS for the Hong Kong population norm. This holds true for both isolated and multiple trauma, and also for each body region studied. At 12 months, only 19% cases had reached a high functional outcome (i.e. GOSE 7), although a further 11% had upper moderate disability (i.e. GOSE = 6). This compares poorly with data from Utrecht in the Netherlands where 79.9% cases actual returned either to part or to full-time work [34] and to Victoria in Australia where 57% returned either to part or to full-time work. Improvements in the MCS are such that in over half of cases they exceed that of the Hong Kong population norm for all injury groups by 6 months. It is possible that the awareness that they have survived significant trauma has a very positive effect, leaving patients with a sense of mental well-being that is greater than the population as a whole. The one group – spinal trauma – with a lower MCS at one year may be due to an awareness of prolonged and possibly non-recoverable injury. After univariate analysis, the significant predictors of long-term (12-month), poor quality of life were age greater than 65 years, male gender, pre-injury health problems, admission to ICU, ISS, baseline PCS, one-month PCS and MCS, and 6-month PCS and MCS. The successful follow up rate at 12 months in Hong Kong was 59%. At least five attempts were made to contact patients by telephone but without success. However, Hong Kong as an integrated medical system whereby it is easy to determine whether the patient is still alive or not. Although it was not possible to evaluate health status in 41% cases, nevertheless the patient was still alive. The proportion of patients with upper good recovery and upper moderate disability at one year after injury is less than those reported from Victoria, Australia [19]. Although we found similar predictors of poorer functional outcome as other studies, such as higher age and comorbidities, nevertheless there were significant differences. For example, female gender did not predict a worse outcome. These differences may result partly from the fact that the Australian population includes only major injury, whilst the Hong Kong population includes both moderate and major injury. Further direct comparison between such centres allowing for important variable interaction would facilitate a useful evaluation of the effectiveness of trauma care. Although the sample size is relatively small, and collected by convenience rather than consecutively, nevertheless this is a landmark study for quality of survival in Asia and for Chinese
patients. The successful follow up rates are less than those reported in more recently published studies [19] but are still of a high level. We did not study the effect of socioeconomic status, educational level or compensation all of which may affect of recovery. The low follow up rate and the reasons for subjects not responding is always a concern in such studies. People may not respond for many extremely different reasons. These include death on the one hand, and excellent recovery on the other. Those with excellent recovery may drop out because they do not wish to go through the trouble of further assessment. Thus there is a risk that the analysis has an inherent systematic bias that among those who did not complete follow up evaluation, there was a higher percentage of excellent recovery. This could be explained by the well recovered patient has less incentive to maintain contact with health care providers. How are we to sort out the impact of these two opposite but potential systematic biases? As already mentioned, Hong Kong has a robust, centralised healthcare database of all patients attending public hospitals. From this database it is possible to follow the survival, reattendance, readmission and clinic records of all cases. Although data from these records are not designed for our study, nevertheless much can be gleaned about the crude survival and functional state of the patients. Data in Table 2 shows the proportion of non-responders at 12 months after injury that were still alive, and therefore at least death is not the reason in the 163 (73%) surviving non-responders. One of the challenges in a study of this nature, is that readers and stakeholders may have different perspectives, and the answers to their questions are affected by the proportion of missing data. Some readers may want to know the change in recovery from baseline, whilst others are interested in specific regional injuries. Thus it is not always clear whether data should be best analysed with reference to the baseline set, or to the available follow up set. The issue could be clearer if each separate analysis was presented as a separate table, but this would increase the number of tables to 12 or 14, which would be unacceptable to the journal. Where possible we have tried to clarify the issue by clearly presenting raw data as numerators and denominators, and giving a clear indication of the primary reference which is stated as ‘100%’. This reference may change depending on whether the reference is the original baseline data, the available data at 1, 6 or 12 month follow up, and whether the analysis is for the dataset for all injuries or for focussed regional injury. As we have presented raw data, then readers with a different perspectives may perform alternative analyses if they wish to obtain percentages that answer their particular question. After the primary analysis of the data, a sensitivity analysis was considered. However, as 12-month responders and surviving nonresponders have similar baseline characteristics, it is reasonable to assume that the proportion of surviving non-responders exceeding the HK norm will be similar to the proportion of responders exceeding the HK norm. The reasons why people may not respond are manifold, but probably includes some who are too well and cannot be bothered to respond, and those patients who are too ill and unable to respond. As contact was not possible, it is not possible to say which if any group predominates. What can reasonably be determined is that death is an unlikely reason for no response, and the baseline characteristics in terms of age, gender, pre-existing morbidity, anatomical and physiological derangement between responders and non-responders are very similar. With this in mind, the conclusions of this study reflect our view that as surviving non-responders have similar baseline characteristics to surviving responders, and the proportion of surviving nonresponders to surviving responders is almost 1:1, then the final recovery of the whole population studied may be double that presented for surviving responders alone.
T.H. Rainer et al. / Injury, Int. J. Care Injured 45 (2014) 902–909
It is important to note that improvements in quality of survival do not appear to have reached a plateau at one year after injury and further study is required for longer term effects. Conclusions For patients sustaining moderate or major trauma in Hong Kong at 12 months after injury less than 1 in 10 patients had an excellent recovery, less than or equal to 3 in 10 patients reached a physical health status score greater or equal to the Hong Kong population norm, although as many as 6 in 10 patients had a mental health status score which is greater or equal to the Hong Kong norm. Conflict of interest statement None. Acknowledgments We thank the Health and Health Services Research Grant Committee of Hong Kong Government for their generous support of this study (HHSRF 07080261). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.injury.2013.11.006. References [1] Sethi D, Aljunid S, Saperi SB, Clemens F, Hardy P, Elbourne D, et al. Comparison of the effectiveness of trauma services provided by secondary and tertiary hospitals in Malaysia. Ann Emerg Med 2007;49:52–61. [2] Tinkoff GH, O’Connor RE, Alexander EL, Jones MS. The Delaware trauma system: impact of level III trauma centers. J Trauma 2007;63:121–7. [3] Rainer TH, Smit P, de V. Trauma systems and emergency medicine. Emerg Med 2003;15:11–7. [4] Cheung NK, Yeung JHH, Chan JTS, Cameron PA, Graham CA, Rainer TH. Primary trauma diversion: initial experience in Hong Kong. J Trauma 2006;61:954–60. [5] Leung GK, Chang A, Cheung FC, Ho HF, Ho W, Hui SM, et al. The first 5 years since trauma center designation in the Hong Kong Special Administrative Region, People’s Republic of China. J Trauma 2011;70:1128–33. [6] Cheng CH, Graham CA, Gabbe BJ, Yeung JH, Kossmann T, Judson RT, et al. Trauma care systems: a comparison of trauma care in Victoria, Australia and Hong Kong, China. Ann Surg 2008;247(2):335–42. [7] Moi AL, Wentzel-Larsen T, Salemark L, Wahl AK, Hanestad BR. Impaired generic health status but perception of good quality of life in survivors of burn injury. J Trauma 2006;61:961–8. [8] Kiely JM, Brasel KJ, Weidner KL, Guse CE, Weigelt JA. Predicting quality of life six months after traumatic injury. J Trauma 2006;61:791–8. [9] Graham CA, Gordon MWG, Roy CW, Hanion PW. Long term outcomes of major trauma without head injury in the West of Scotland: pilot case control study. Eur J Emerg Med 2007;14:35–8. [10] Mann N, Mullins R. Research recommendations and proposed action items to facilitate trauma system implementation and evaluation. J Trauma 1999;47: S75–8.
909
[11] Utomo WK, Gabbe BJ, Simpson PM, Cameron PA. Predictors of in-hospital mortality and 6-month functional outcomes in older adults after moderate to severe traumatic brain injury. Injury 2009;40:973–7. [12] Gabbe BJ, Sutherland AM, Williamson OD, Cameron PA. Use of health care services 6 months following major trauma. Aust Health Rev 2007;31:628–32. [13] O’Mullane PA, Mikocka-Walus AA, Gabbe BJ, Cameron PA. Incidence and outcomes of major trauma assaults: a population-based study in Victoria. Med J Aust 2009;190:129–32. [14] Gruen RL, Gabbe BJ, Stelfox HT, Cameron PA. Indicators of the quality of trauma care and the performance of trauma systems. Br J Surg 2012;99(Suppl. 1):97– 104. http://dx.doi.org/10.1002/bjs.7754. [15] Gabbe BJ, Simpson PM, Sutherland AM, Palmer CS, Williamson OD, Butt W, et al. Functional and health-related quality of life outcomes after pediatric trauma. J Trauma 2011;70:1532–8. [16] Williamson OD, Gabbe BJ, Sutherland AM, Wolfe R, Forbes AB, Cameron PA. Comparing the responsiveness of functional outcome assessment measures for trauma registries. J Trauma 2011;71:63–8. [17] Nichol AD, Higgins AM, Gabbe BJ, Murray LJ, Cooper DJ, Cameron PA. Measuring functional and quality of life outcomes following major head injury: common scales and checklists. Injury 2011;42:281–7. [18] MacKenzie EJ, Rivara FP, Jurkovitch GJ, Nathens AB, Egleston BL, Slaveker DS, et al. The impact of trauma-center care on functional outcomes following major lower limb trauma. J Bone Joint Surg 2008;90-A:101–9. [19] Gabbe BJ, Simpson PM, Sutherland AM, Wolfe R, Fitzgerald MC, Judson R, et al. Improved functional outcomes for major trauma patients in a regionalized, inclusive trauma system. Ann Surg 2012;255:1009–15. [20] Gabbe BJ, Sleney J, Gosling C, Wilson K, Sutherland A, Hart M, et al. Exploring trauma patient perception of barriers and facilitators to recovery. In: Conference Proceedings of the Annual Scientific Meeting of the Australasian Trauma Society. Trauma 2012: Resuscitation to Rehabilitation; 2012. p. A35. [21] Lam CL, Lauder IJ, Lam TP, Gandek B. Population based norming of the Chinese (HK) version of the SF-36 health survey. Hong Kong Pract 1999;21: 460–70. [22] Lam CL, Tse EYY, Gandek B, Fong DY. The SF-36 summary scales were valid, reliable, and equivalent in a Chinese population. J Clin Epidemiol 2005;58:815–22. [23] Bowling A. Measuring health: a review of quality of life measurement scales. Buckingham: Open University Press; 1997. [24] Garratt A, Schmidt L, Mackintosh A, Fitzpatrick R. Quality of life measurement: bibliographic study of patient assessed health outcome measures. Br Med J 2002;324:1417. [25] Nijboer JMM, van der Sluis CK, van der Naalt J, Nijsten MW, Ten Duis HJ. Two cohorts of severely injured trauma patients, nearly two decades apart: unchanged mortality but improved quality of life despite higher age. J Trauma 2007;63:670–5. [26] Wilson JTL, Pettigrew LEL, Teasdale GM. Structured interviews for the Glasgow Outcome Scale and Extended Glasgow Outcome Scale: guidelines for their use. J Neurotrauma 1998;15:573–85. [27] Brenneman FD, Redelmeier DA, Boulanger BR, McLellan BA, Culhane JP. Longterm outcomes in blunt trauma: who back to work? J Trauma 1997;42: 778–81. [28] MacKenzie EJ, McCarthy ML, Ditunno JF, Forrester-Staz C, Gruen GS, Marion DW, et al. Using the SF36 for characterising outcome after multiple trauma involving head injury. J Trauma 2002;52:527–34. [29] Baker SP, O’Neill B, Haddon Jr W, Long WB. The Injury Severity Score: a method for describing patients with multiple injuries and evaluating emergency care. J Trauma 1974;14:187–96. [30] Association for the advancement of automative medicine. In: The abbreviated injury scale. Des Plaines Ilinois; 1990 Revision. [31] Champion HR, Sacco WJ, Copes WS, Gann DS, Gennarelli TA, Flanagan ME. A revision of the Trauma Score. J Trauma 1989;29(5):623–9. [32] Boyd CR, Tolson MA, Copes WS. Evaluating trauma care: the TRISS method. Trauma Score and the Injury Severity Score. J Trauma 1987;27:370–8. [33] http://www.qualitymetric.com/ [accessed 10.08.12]. [34] Holtslag HR, Post MW, van der Werken C, Lindeman E. Return to work after major trauma. Clin Rehabil 2007;21:373–83.
