Accepted Manuscript Title: Plasma melatonin levels in hip fracture patients with and without delirium: A confirmation study Authors: Rikie M. Scholtens, Barbara C. van Munster, Martijn van Faassen, Marijn F. van Kempen, Ido P. Kema, Sophia E. de Rooij PII: DOI: Reference:
S0047-6374(17)30005-2 http://dx.doi.org/10.1016/j.mad.2017.08.016 MAD 10990
To appear in:
Mechanisms of Ageing and Development
Received date: Revised date: Accepted date:
29-12-2016 31-7-2017 22-8-2017
Please cite this article as: Scholtens, Rikie M., van Munster, Barbara C., van Faassen, Martijn, van Kempen, Marijn F., Kema, Ido P., de Rooij, Sophia E., Plasma melatonin levels in hip fracture patients with and without delirium: A confirmation study.Mechanisms of Ageing and Development http://dx.doi.org/10.1016/j.mad.2017.08.016 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
2135 Words (excl references, abstract and legends) 3 Tables 0 Figures 27 References
Plasma melatonin levels in hip fracture patients with and without delirium: a confirmation study Rikie M. Scholtens, MD*ac, Barbara C. van Munster, MD, PhDbc, Martijn van Faassen, BAScd, Marijn F. van Kempen, MDe , Ido P. Kema, PhDd, Sophia E. de Rooij, MD, PhDac
a
Department of Internal Medicine, Geriatrics Section, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands b
Department of Geriatrics, Gelre Hospitals, Apeldoorn, The Netherlands
c
University Center for Geriatric Medicine, University Medical Center Groningen, Groningen, The Netherlands d
Department of Laboratory Medicine, University Medical Center Groningen, Groningen, The Netherlands e
Leiden University Medical Center, Leiden University, Leiden, The Netherlands
*Corresponding author: Rikie Scholtens, Academic Medical Center, University of Amsterdam, Department of Internal Medicine, Geriatrics section, P.O. Box 22660, 1100 DD, Amsterdam, The Netherlands. Phone: +31 20 5661647 Fax: + 31 20 5669325 E-mail:
[email protected]
Highlights Melatonin plays a major role in maintaining circadian rhythm Sleep and wake cycle disturbances are important features of patients with delirium Measuring plasma melatonin could provide more insight into delirium pathophysiology High morning melatonin levels were associated with age and postoperative samples 1
Abstract
Background: Melatonin plays a major role in maintaining circadian rhythm. Changes in melatonin metabolism might lead to circadian rhythm disturbances which are often observed in delirious patients. Aim: To assess if high morning plasma melatonin concentrations were associated with delirium. Methods: Consecutive hip fracture patients aged ≥65 years were included. Delirium was assessed daily with the Confusion Assessment Method. Blood samples were collected at 11.00am on weekdays during first week of hospitalization. Melatonin was analyzed by liquid chromatography-tandem mass spectrometry. Results: We analyzed 389 samples of 144 participants [mean age 84.0, 70 experienced delirium]. A Generalized Estimating Equations (GEE) model with outcome melatonin level in highest tertile (>3.36 pg/ml) and covariates delirium group (i.e. never, before, during, post delirium), cognitive impairment, age, sex and anesthesia type, was constructed. Highest melatonin levels were associated with postoperative samples (Odds Ratio(OR) 2.11 compared to preoperative samples; 95% Confidence Interval(CI) 1.17–3.82, p=0.01) and higher age (OR 1.05 per year; CI 1.01-1.11, p=0.03), but not with delirium group(p=0.35). Conclusion: Undergoing surgery and aging in general may induce changes in melatonin metabolism. Future research should focus on daily multiple melatonin measurements to determine whether melatonin supplementation might be beneficial for delirium treatment or prevention. Key words: Delirium, melatonin, circadian rhythm, hip fracture, aged, elderly 2
1. Introduction
Delirium is an acute neuropsychiatric syndrome that particularly affects older people. Higher age, undergoing surgery and sleep deprivation are factors known to precipitate delirium. (1) It is associated with negative long term outcomes like institutionalization, cognitive impairment and even death. (2) The pathophysiology of delirium has not been clarified yet. Delirium is characterized by changes in cognition, disturbances in consciousness and the possible development of a perceptual disturbance. (3) Disturbances in sleep and wake cycle and changes in (motor) activity are also important features of patients with delirium (4) and could be considered as circadian rhythm disturbances. The pineal hormone melatonin plays a major role in maintaining circadian rhythm. Melatonin is produced in absence of light and maximal levels are reached between 01:00am and 04:00am. (5) One could hypothesize that changes in melatonin metabolism, like decreased production or a phase shift, might lead to, or attribute to, circadian rhythm disturbances which are often seen in patients with postoperative delirium. (6) Measuring plasma melatonin levels in patients at high risk for delirium could therefore provide more insight into delirium pathophysiology. Precipitating factors for delirium like surgery and cognitive impairment could lower nocturnal melatonin concentrations or change secretion patterns, suggesting that these events might induce greater vulnerability for circadian rhythm disturbances. (7-9) Previous research on melatonin levels in patients with delirium showed the maximal nocturnal melatonin concentrations to decline during delirium, and that daytime melatonin levels also could be higher in delirious patients compared to physiological daytime melatonin levels in nondelirious subjects. (10-12) So far, melatonin supplementation for delirium prevention has shown contradictory results. (13) A possible explanation for these contradictory results could 3
be that not only a melatonin production deficit plays a role in delirium pathophysiology, but also a derailed timing of maximal levels. Declined nocturnal melatonin levels and higher daytime levels might indicate that maximal nocturnal levels would be reached at a different time point, and might lead to higher daytime levels (phase shift). Previously, we measured perioperative plasma melatonin levels in older hip fracture patients and found higher morning plasma melatonin levels to be associated with samples obtained post-operatively compared to samples obtained pre-surgery, (Odds Ratio (OR) 2.55, 95% Confidence Interval (CI) 1.52–4.30, p<0.01), but we did not find an association with delirium. (14) To further validate these findings, we have performed the present study and analyzed morning plasma melatonin levels in another cohort of elderly hip fracture patients.
2. Methods 2.1 Design, setting and participants This is a post hoc analysis of a cohort study investigating delirium pathophysiology. From May 2005 to February 2008, consecutive patients aged 65 years and older who were acutely admitted for surgical repair of a hip fracture to the Department of Trauma Surgery or Orthopaedic Surgery of the Academic Medical Center in Amsterdam, the Netherlands, were invited to participate. Informed consent was obtained from all patients or their legal representatives in cases of cognitive impairment. The institutional Medical Ethics Committee approved the study. 2.2 Data collection At baseline we collected demographic data, data of the surgical procedure and number of drugs used at home. Comorbidities were scored with the Charlson Comorbidity Index (CCI). 4
(15) The 15-item Katz Activities of Daily Living (ADL) scale (16) was completed by patients or relatives to explore pre-existing physical functionality. We assessed premorbid cognitive functioning with the Informant Questionnaire on Cognitive Decline short form (IQCODE-sf) (17) and asked patients to compare the present situation with two weeks prior to hip fracture. A cut-off score of ≥ 3.9 on this scale or a diagnosis of dementia in the medical history indicated global cognitive impairment. 2.3 Delirium assessments Delirium was evaluated daily on weekdays by a geriatrician and a research nurse with experience in geriatrics using the Confusion Assessment Method (CAM) (18). Additionally, we performed a psychiatric examination and used data from medical and nursing records, including the Delirium Observation Screening Scale (DOSS) (19) and information given by the patient’s closest relative to diagnose possible delirium. 2.4 Blood sample collection and melatonin measurement A maximum of four blood samples was collected at 11.00am on weekdays during the first week of hospitalization. Blood was collected in tubes containing ethylenediaminetetraacetic acid (EDTA) and kept on ice after withdrawal. Plasma was obtained using centrifugation for 15 minutes at 1,780g at 4°C. Aliquots have been stored at -80°C since collection and have gone through several freeze and thaw cycles before melatonin measurement was performed. Melatonin was analyzed by online-solid phase extraction in combination with isotope dilution liquid chromatography-tandem mass spectrometry (LC-MS/MS). Mean intra- and inter-assay coefficients of variation were below 10%. Melatonin concentrations below the quantification limit of 1.16 pg/ml (5.0 pmol/L) were set at half the value. 2.5 Statistical analyses 5
For calculating differences in baseline characteristics we used a t-test, chi-square test or Mann Whitney U test, as appropriate. We allocated samples to preoperative samples or postoperative samples. Samples were categorized into four groups: 1] obtained before delirium, 2] during delirium, 3] post delirium, and 4] samples of patients who did not experience delirium during hospitalization. Samples obtained until the day before delirium diagnosis were allocated to ‘before delirium’. Samples obtained from the first day without delirium were allocated to ‘post delirium’. If delirium assessment was missing or if a sample was taken in between two days with delirium, we excluded these from the analysis. Melatonin concentrations were divided into tertiles(T). We defined high melatonin levels as levels in the highest tertile T3. A Generalized Estimating Equations(GEE) model with binary outcome T3 or
0.05 and the Corrected Quasi Likelihood under Independence Model Criterion (QICC) improved. All statistical analyses were performed in SPSS 23.0.
3. Results 3.1 Descriptive statistics Three hundred and thirteen consecutive older hip fracture patients underwent hip surgery in the study period. Of these, 30 patients met the exclusion criteria, 60 patients denied participation and of 79 patients no blood sample was available. In the present analyses we included 144 participants of whom 70 (48.6%) experienced delirium during hospitalization. 6
Patients who developed delirium were older, more often institutionalized, had more often cognitive and functional impairment, and had more comorbidities. (Table 1) 3.2 Melatonin concentrations We analyzed 389 samples of which 33 were obtained before delirium, 117 during delirium, 43 post delirium, and 196 of patients who were not delirious during hospitalization. We divided melatonin concentrations into T1 <0.58 pg/ml, T2 0.59 – 3.36 pg/ml and T3 >3.36 pg/ml. In univariate analysis presence of premorbid cognitive impairment (OR 1.79, 95% CI 0.99-3.25 p=0.05), higher age (OR 1.06 per year 95% CI 1.01-1.11 p=0.01), postoperative samples (OR 1.86 95% CI 1.13-3.09 p=0.02) and male sex (OR 0.53 95% CI 0.28-1.01 p=0.05) were associated with melatonin levels in T3. (Table 2) A multivariable model with age, preoperative or postoperative sample, delirium group and sex resulted in the best model fit. (Table 3) GEE showed an OR of 2.11 on highest melatonin levels in postoperative samples compared to preoperative samples (95% CI 1.17–3.82, p=0.01) and an OR of 1.05 for each year increase in age (95% CI 1.01-1.11, p=0.03). Highest melatonin levels were no longer associated with sex or cognitive impairment, nor with delirium group.
4. Discussion
In this post hoc analysis of a study on 144 older hip fracture patients with a high risk on delirium, we found high morning plasma melatonin levels to be associated with postoperative samples and higher age, but not with delirium. These findings are in line with our previous study on morning melatonin levels in older hip fracture patients, in which we were not able to demonstrate an association between higher melatonin levels and delirium. (14) In 30 elderly 7
general medicine inpatients higher daytime melatonin levels were found during delirium compared to post delirium. (10) A possible explanation for these differences could be that in the surgical population more precipitating factors for delirium are present that are also known to influence melatonin metabolism, like undergoing surgery, administering anesthetics and inflammatory processes. (7, 8, 20, 21). Research on postoperative melatonin levels showed lower melatonin levels in the first postoperative nights and changes in secretion patterns compared to pre-surgery. (7, 12) We found higher morning plasma melatonin levels in postoperative samples compared to preoperative samples. One could speculate that high daytime melatonin levels indicate that the secretion pattern of melatonin has been changed, for example by a phase shift, having multiple peak concentrations in 24 hours or continuously high levels. Our findings could indicate that undergoing surgery might induce circadian rhythm disturbances and not a decrease in melatonin production, although further research is needed to confirm this hypothesis. We found an association of high morning melatonin levels with cognitive impairment in the univariate analysis, but not in multivariate analysis. Alterations like low nocturnal plasma melatonin levels, abolished secretion pattern or even patterns in which no relation with day light or an individual’s phase position could be determined (free running) have been described in patients with dementia. These changes in melatonin levels could be a possible explanation for circadian rhythm disturbances (like sundowning) often seen in these patients and might lead to greater vulnerability for developing both melatonin phase shifts and delirium.(9, 2224) However, because of these heterogeneously and possible pre-existent changes in populations with high prevalence of cognitive impairment, detecting differences in melatonin metabolism subscribed to delirium might be difficult. 8
In this study increasing age was associated with high morning plasma melatonin levels. In previous research maximal nocturnal melatonin levels were found to decline with increasing age in normal subjects. Secretion patterns were also shown to differ, which could lead to higher daytime levels. (25, 26). 4.1 Strengths and limitations In this study with a large sample size on vulnerable elderly at high risk for delirium we analyzed 389 samples collected during hospitalization and performed a post hoc analysis on morning plasma melatonin levels. By replicating our previous study on melatonin levels in older hip fracture patients, we were able to validate our findings in another surgical cohort, which attributes to the body of knowledge on delirium pathophysiology in hip fracture patients. Although by collecting one sample per day no daily secretion pattern, maximal levels or dim light melatonin onset (DLMO) could be determined, it can be considered as a first step in exploring contradictory results on melatonin supplementation in diverse populations. Overall, the melatonin concentrations we found in this study were low. A possible explanation could be that samples have been stored at -80°C for several years and have gone through several freeze and thaw cycles, which could have influenced melatonin stability. Also differences in measurement technique could have influenced results. Radioimmunoassay (RIA) and Enzyme-Linked Immuno Sorbent Assay (ELISA) are known to have interference with (N-acetyl-)serotonin and LC-MS/MS seems more specific in detecting melatonin, resulting in lower melatonin levels.(27) Therefore, comparing melatonin levels across studies is difficult and this endorses the need for a gold standard for melatonin measurements. The samples in this trial were not collected with the present hypothesis in mind and therefore information on sleep, light, medication use, detailed medical and psychiatric history, diet and individual chronotype is lacking. Also, no neuroimaging was performed to assess anatomical 9
changes of the pineal gland. Future prospective studies should take these factors into account. Obtaining blood samples in a hospital setting, which is a 24 hours bright lightened environment with often noise at night, and in stressful and perhaps painful circumstances for the patient, could all influence melatonin levels. However, because this affects all patients included in this study equally, we believe this is not of major influence on the associations found. So far, in hip fracture patients effect of melatonin supplementation on delirium duration, but not on delirium incidence was found, in contrast to other populations.(13) The population studied, especially regarding prevalence of cognitive impairment and undergoing surgery, and setting seems to play an important role. Therefore, melatonin secretion patterns must be further explored to assess for which populations melatonin supplementation could be beneficial for delirium prevention or treatment of symptoms of delirium. Future research should therefore focus on daily multiple melatonin measurements to further explore the association of melatonin levels and changes in melatonin secretion patterns within delirium episodes. 4.2 Conclusions Undergoing surgery and aging in general may induce changes in melatonin metabolism which could possibly lead to circadian rhythm disturbances. In this study high morning plasma melatonin levels were not associated with delirium.
Acknowledgements The authors have no conflicts of interest to declare. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
10
References 1. Inouye SK. Predisposing and precipitating factors for delirium in hospitalized older patients. Dementia and geriatric cognitive disorders. 1999;10(5):393-400. 2. Witlox J, Eurelings LS, de Jonghe JF, Kalisvaart KJ, Eikelenboom P, van Gool WA. Delirium in elderly patients and the risk of postdischarge mortality, institutionalization, and dementia: a meta-analysis. JAMA. 2010;304(4):443-51. 3. American Psychiatric Association. Diagnostic and statistical manual of mental disorders 4th edition (DSM-IV-TR). 4th, text revision ed. (APA) APA, editor. Washington, DC, USA: American Psychiatric Press; 2000. 4. Meagher DJ, Moran M, Raju B, Gibbons D, Donnelly S, Saunders J, et al. Phenomenology of delirium. Assessment of 100 adult cases using standardised measures. Br J Psychiatry. 2007;190:135-41. 5. Brzezinski A. Melatonin in humans. N Engl J Med. 1997;336(3):186-95. 6. Fitzgerald JM, Adamis D, Trzepacz PT, O'Regan N, Timmons S, Dunne C, et al. Delirium: a disturbance of circadian integrity? Med Hypotheses. 2013;81(4):568-76. 7. Cronin AJ, Keifer JC, Davies MF, King TS, Bixler EO. Melatonin secretion after surgery. Lancet. 2000;356(9237):1244-5. 8. Miyazaki T, Kuwano H, Kato H, Ando H, Kimura H, Inose T, et al. Correlation between serum melatonin circadian rhythm and intensive care unit psychosis after thoracic esophagectomy. Surgery. 2003;133(6):662-8. 9. Uchida K, Okamoto N, Ohara K, Morita Y. Daily rhythm of serum melatonin in patients with dementia of the degenerate type. Brain Res 1996;717(1-2):154-9. 10. Piotrowicz K, Klich-Raczka A, Pac A, Zdzienicka A, Grodzicki T. The diurnal profile of melatonin during delirium in elderly patients-preliminary results. Exp Gerontol. 2015;72:45-9. 11. Uchida K, Aoki T, Ishizuka B. Postoperative delirium and plasma melatonin. Med Hypotheses. 1999;53(2):103-6. 12. Shigeta H, Yasui A, Nimura Y, Machida N, Kageyama M, Miura M, et al. Postoperative delirium and melatonin levels in elderly patients. Am J Surg 2001;182(5):449-54. 13. Siddiqi N, Harrison JK, Clegg A, Teale EA, Young J, Taylor J, et al. Interventions for preventing delirium in hospitalised non-ICU patients. The Cochrane database of systematic reviews. 2016;3:Cd005563. 14. Scholtens RM, van Munster BC, Maijers I, Maas HAAM, de Jonghe A, de Rooij SE. Higher morning plasma melatonin levels are not associated with delirium. submitted. 2016. 15. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987;40(5):373-83. 16. Weinberger M, Samsa GP, Schmader K, Greenberg SM, Carr DB, Wildman DS. Comparing proxy and patients' perceptions of patients' functional status: results from an outpatient geriatric clinic. J Am Geriatr Soc. 1992;40(6):585-8. 17. Jorm AF. The Informant Questionnaire on cognitive decline in the elderly (IQCODE): a review. International psychogeriatrics / IPA. 2004;16(3):275-93. 18. Inouye SK, van Dyck CH, Alessi CA, Balkin S, Siegal AP, Horwitz RI. Clarifying confusion: the confusion assessment method. A new method for detection of delirium. Annals of internal medicine. 1990;113(12):941-8. 19. Schuurmans MJ, Shortridge-Baggett LM, Duursma SA. The Delirium Observation Screening Scale: a screening instrument for delirium. Research and theory for nursing practice. 2003;17(1):31-50. 20. Reber A, Huber PR, Ummenhofer W, Gurtler CM, Zurschmiede C, Drewe J, et al. General anaesthesia for surgery can influence circulating melatonin during daylight hours. Acta Anaesthesiol Scand. 1998;42(9):1050-6. 21. Bourne RS, Mills GH. Melatonin: possible implications for the postoperative and critically ill patient. Intensive Care Med. 2006;32(3):371-9. 22. Ohashi Y, Okamoto N, Uchida K, Iyo M, Mori N, Morita Y. Daily rhythm of serum melatonin levels and effect of light exposure in patients with dementia of the Alzheimer's type. Biol Psychiatry. 1999;45(12):1646-52. 23. Mishima K, Tozawa T, Satoh K, Matsumoto Y, Hishikawa Y, Okawa M. Melatonin secretion rhythm disorders in patients with senile dementia of Alzheimer's type with disturbed sleep-waking. Biol Psychiatry. 1999;45(4):417-21. 24. Skene DJ, Swaab DF. Melatonin rhythmicity: effect of age and Alzheimer's disease. Exp Gerontol 2003;38(1-2):199-206. 11
25. Zhao ZY, Xie Y, Fu YR, Bogdan A, Touitou Y. Aging and the circadian rhythm of melatonin: a crosssectional study of Chinese subjects 30-110 yr of age. Chronobiol Int 2002;19(6):1171-82. 26. Scholtens RM, van Munster BC, van Kempen MF, de Rooij SE. Physiological melatonin levels in healthy older people: A systematic review. Journal of psychosomatic research. 2016;86:20-7. 27. de Almeida EA, Di Mascio P, Harumi T, Spence DW, Moscovitch A, Hardeland R, et al. Measurement of melatonin in body fluids: standards, protocols and procedures. Childs Nerv Syst. 2011;27(6):879-91.
Table 1. Characteristics of participants Variable
Delirium
No delirium
pvalue
N=70
N=74
Mean age in years (SD)
85.5 (6.6)
82.5 (7.0)
0.01
Sex, male
18 (25.7)
23 (31.1)
0.48
Cognitive impairment
44 (62.9)
11 (14.9)
<0.01
Living at home
35 (50.0)
59 (79.7)
<0.01
Median Charlson Comorbidity Index (IQR)
6 (2.0)
4 (3.0)
<0.01
Median preadmission Katz-15-ADL (IQR)
8 (5.0)
4 (5.0)
<0.01
missing (%)
3 (4.3)
2 (2.7)
Median number of drugs used at home (IQR)
4.5 (5.0)
4 (5.0)
missing (%)
16 (22.9)
21 (28.4)
Median length of hospital stay in days (IQR)
12.5 (15.3)
11.0 (9.0)
0.16
Median time to surgery in days (IQR)
1.0 (0.25)
1.0 (1.0)
0.74
0.13
0.26
Type of anesthesia spinal anesthesia
19 (27.1)
25 (33.8)
general anesthesia
49 (70.0)
49 (66.2)
other
2 (2.9)
0 (0.0) 0.64
Type of fracture femoral neck
32 (45.7)
35 (47.3)
intertrochanteric
34 (48.6)
32 (43.2)
12
other
4 (5.7)
7 (9.5) 0.69
Type of surgery hip replacement
27 (38.6)
31 (41.9)
internal fixation
43 (61.4)
43 (58.1)
All characteristics are shown as number (%) unless stated otherwise. Katz-15-ADL: 15-item Katz Index of Activities of Daily Living SD: Standard Deviation IQR: Inter Quartile Rate
Table 2. Univariate GEE parameter estimates for plasma melatonin concentrations in highest tertile (> 3.36 pg/ml) of all parameters of interest.
Parameter
B
S.E.
Before delirium
0.63
0.39
Wald ChiSquare 2.64
p-value
Odds ratio
0.10
1.87
95% Confidence Interval 0.88 – 3.99
During delirium
0.31
0.31
1.00
0.32
1.37
0.74 – 2.51
Post delirium
0.53
0.38
1.96
0.16
1.71
0.81 – 3.60
No delirium
0a
Cognitive impairment
0.58
No cognitive impairment
0a
General anesthesia
0.08
Spinal anesthesia
0a
Age in years
0.06
0.02
6.45
0.01
1.06
1.01 – 1.11
Male sex
-0.64
0.33
3.75
0.05
0.53
0.28 – 1.01
Female sex
0a
Postoperative sample
0.62
Preoperative sample
0a
1.00 0.30
3.69
0.05
1.79
0.99 – 3.25
1.00 0.33
0.07
0.80
1.09
0.57 – 2.08
1.00
1.00 0.26
5.85
0.02
1.86
1.13 – 3.09
1.00
a: reference category B: parameter estimate S.E.: standard error Samples were obtained at 11:00 am.
13
Table 3. Parameter estimates of the final multivariate GEE for plasma melatonin levels in highest tertile (> 3.36 pg/ml) at 11:00 am. Wald Odds Chip-value ratio Square 2.64 0.10 0.57
95% Confidence Interval
Parameter
B
S.E.
Male sex
-0.56
0.35
Female sex
0a
Before delirium
0.59
0.41
2.11
0.15
1.81
0.81 – 4.02
During delirium
0.07
0.33
0.04
0.84
1.07
0.56 – 2.04
Post delirium
0.29
0.43
0.45
0.50
1.33
0.58 – 3.08
No delirium
0a
Postoperative sample
0.75
Preoperative sample
0a
Age in years
0.05
0.29 – 1.12
1.00
1.00 0.30
6.16
0.01
2.11
1.17 – 3.82
1.00 0.02
4.79
0.03
1.05
1.01 – 1.11
a: reference category B: parameter estimate S.E.: standard error
14