Small Business & Entrepreneurship

A Multifactorial Intervention Program Reduces the Duration of Delirium, Length of Hospitalization, and Mortality in Delirious Patients

Description
To investigate whether an education program and a reorganization of nursing and medical care improved the outcome for older delirious patients. Prospective intervention study. Department of General Internal Medicine, Sundsvall Hospital, Sweden. Four
Published
of 7
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  A Multifactorial Intervention Program Reduces the Durationof Delirium, Length of Hospitalization, and Mortality inDelirious Patients Maria Lundstro¨ m, RN, PhD,  Agneta Edlund, RN, w  Stig Karlsson, RN, PhD,  Benny Bra¨nnstro¨ m, RN, PhD, z  Go¨ sta Bucht, MD, PhD,  and Yngve Gustafson, MD, PhD  OBJECTIVES:  To investigate whether an education pro-gram and a reorganization of nursing and medical care im-proved the outcome for older delirious patients. DESIGN:  Prospective intervention study. SETTING:  Department of General Internal Medicine,Sundsvall Hospital, Sweden. PARTICIPANTS:  Fourhundredpatients,aged70andolder,consecutively admitted to an intervention or a control ward. INTERVENTION:  The intervention consisted of staff ed-ucation focusing on the assessment, prevention, and treat-ment of delirium and on caregiver-patient interaction.Reorganization from a task-allocation care system to a pa-tient-allocation system with individualized care. MEASUREMENTS:  The patients were assessed using theOrganic Brain Syndrome Scale and the Mini-Mental StateExamination on Days 1, 3, and 7 after admission. Deliriumwas diagnosed according to  Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition,  criteria. RESULTS:  Delirium was equally common on the day of admission at the two wards, but fewer patients remaineddelirious on Day 7 on the intervention ward (n 5 19/63,30.2% vs 37/62, 59.7%,  P 5 .001). The mean length of hospital stay  standard deviation was significantly loweron the intervention ward then on the control ward(9.4  8.2 vs 13.4  12.3 days,  P o .001) especially for thedeliriouspatients(10.8  8.3vs20.5  17.2days, P o .001).Two delirious patients in the intervention ward and nine inthe control ward died during hospitalization ( P 5 .03). CONCLUSION:  This study shows that a multifactorialintervention program reduces the duration of delirium,length of hospital stay, and mortality in delirious patients.  J Am Geriatr Soc 53:622–628, 2005. Key words: delirium; intervention; multifactorial; lengthof hospitalization; mortality D elirium, as defined in the  Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition  (DSM-IV), is a common neuropsychiatric syndrome characterizedbydisturbanceinattentionandconsciousnessthatdevelopsover a short period oftime and in which thesymptoms tendto fluctuate during the course of the day. 1 Delirium is prob-ably the most common presenting symptom of disease inold age.Delirium occurs in 14% to 42% of older peopleadmitted to general internal medicine or acute geriatricunits 2–8 andin28%to61%ofolderpatientstreatedforhipfractures. 9–11 In general internal medicine or acute geriatricunits, delirium has been reported to be associated with oldage, dementia, stroke, myocardial infarction, pneumonia,and electrolyte disturbances. 5,12,13 Delirium has also beenreported to be associated with prolonged hospitalization 2,8 increased mortality, 14,16 and the development of demen-tia. 15,17,18 Several intervention studies have shown that postoper-ative delirium after hip-fracture surgery can be preventedand treated. 19–23 Successful intervention programs are mul-tifactorial and interdisciplinary and include the assessmentandtreatmentofunderlyingcausesaswellasthepreventionand treatment of factors endangering the cerebral metab-olism. Excellent nursing care seems to be a prerequisite forsuccessful intervention, according to the majority of thestudies. 19,21,22 Few intervention studies have been performed in gen-eral internal medicine patients, and only one has shownany positive results in reducing delirium. 24–27 The success-ful intervention study was a multicomponent risk-factor ThestudywassupportedbygrantsfromtheJointCommitteeoftheNorthernHealth Region of Sweden (Visare Norr), the Gun and Bertil Stohne’sFoundation, the Borgerskapet of Umea˚ Research Foundation, the JC KempeMemorial Foundation, the Foundations of the Medical Faculty, Universityof Umea˚, the Federation of County Councils (Dagmar), and the Detlof’sFoundation.Address correspondence to Maria Lundstro ¨m, Department of CommunityMedicine and Rehabilitation, Geriatric Medicine, Umea˚ University, SE-90187 Umea˚, Sweden. E-mail: maria.lundstrom@germed.umu.seFrom the   Department of Community Medicine and Rehabilitation,Geriatric Medicine, Umea˚ University, Umea˚, Sweden;  w Department of Medicine and Rehabilitation, Pitea˚ River Valley Hospital, Pitea˚, Sweden; and z Department of Health Sciences, Lulea˚ University of Technology, Boden,Sweden.  JAGS 53:622–628, 2005 r 2005 by the American Geriatrics Society  0002-8614/05/$15.00  intervention study with the aim of preventing the develop-ment of delirium in older patients and resulted in a reduc-tion in the number and duration of episodes of delirium. 25 To the authors’ knowledge, no intervention study inolder patients admitted to general internal medicine hassucceeded in reducing the duration of delirium and thus thelength of hospitalization in patients with delirium on ad-mission to general internal medicine wards.The aim of the study was to investigate whether aneducation program and a reorganization of nursing andmedicalcarewouldimprovetheoutcomeforolderdeliriouspatients admitted to general internal medicine wards. METHODSStudy Population Four hundred patients, aged 70 and older, consecutivelyadmitted to two wards (one intervention and one controlward, 200 to each ward) in the Department of GeneralInternal Medicine at Sundsvall Hospital, Sweden, over an8-month period were included in the study. The only ex-clusion criteria were age younger than 70 and patient’s re-fusal to participate. All other patients were includedregardless of diagnosis. Patient characteristics are present-ed in Table 1. Study Procedures The patients were mainly (93.8%) admitted from the emer-gency room in the same proportion to each ward. The pa-tients were randomly allocated to any ward with anaccessible bed, but patients who were readmitted (n 5 23)within 3 months after their latest discharge were, if possi-ble, admitted to the ward where they had been cared for onthe earlier occasion. One of the wards was a subspecialistward in endocrinology, and their planned admissions weremainly patients with complications from diabetes mellitus.All other patients were included regardless of diagnosis.Thepatientsgavetheirconsenttoparticipateinthestudyatthe first assessment occasion on the ward during the firstday of admission. The staff on the intervention ward werefully aware of the nature of the study, and the staff on thecontrol ward were informed that a screening of the prev-alence of delirium was being performed. Intervention The program on the intervention ward consisted of fourparts:1. A 2-day course for staff on geriatric medicine focusingon assessment, prevention, and treatment of delirium2. Education concerning caregiver-patient interaction fo-cusing on patients with dementia and delirium3. Reorganization from a task-allocation care system to apatient-allocation system with individualized care4. Guidance for nursing staff once a month Course in Geriatric Medicine Focusing on Delirium All members of the nursing and medical staff attended a 2-day course in geriatric medicine and nursing focusing ondementia and delirium and its risk factors, treatment, andcare. These first basic lectures took place before the start of the study, with a follow-up during the first month of thestudy. The program was mainly based on two doctoral the-ses by two of the authors (YG, BB). The training regardingmedical intervention focused on prevention of hypoxemia,hypercortisolism, and other factors that could endanger thecholinergicmetabolisminthepatients’brain,suchasavoid-ing drugs with anticholinergic properties. 11,23,28 The train-ing also focused on assessment and treatment of underlyingcauses of delirium, such as urinary tract infection, pneu-monia, constipation, heart failure, pulmonary embolism,and drug side effects. The training regarding nursing inter- Table 1. Characteristics and Previous Diagnoses of All Patients and of Delirious Patients on the Intervention and ControlWards CharacteristicAll Patients Delirious PatientsIntervention Ward Control Ward P  -valueIntervention Ward Control Ward P  -value(n 5 200) (n 5 200) (n 5 63) (n 5 62)Male/female, % 39.0/61.0 49.5/50.5 .04 44.4/55.6 61.3/38.7 .06Age, mean  standarddeviation79.4  5.6 80.7  6.2 .02 81.6  6.0 81.9  6.6 .76Impaired hearing(n 5 198/198 vs 61/61), %1.5 4.0 .13 3.3 9.8 .27Impaired vision(n 5 193/189 vs 57/55), %15.0 17.0 .61 19.3 34.6 .07Asthma, % 13.0 10.5 .44 7.9 12.9 .36Dementia, % 4.5 4.5 1.00 4.8 6.4 .72Diabetes mellitus, % 42.5 23.5  o .001 39.7 19.4 .01Epilepsy, % 4.5 4.5 1.00 11.1 4.8 .32Malignancies, % 9.0 13.0 .20 6.4 8.1 .71Stroke, % 25.5 24.0 .73 39.7 32.2 .39Admitted from ownhouse/department withor without home-help service, %82.0 88.0 .09 73.0 80.6 .31 MULTIFACTORIAL DELIRIUM INTERVENTION STUDY  623 JAGS APRIL 2005–VOL. 53, NO. 4  vention focused on interaction with patients with reducedattention and orientation in a stressful situation and how tooptimize how care could be provided to such patients. 29 The organization of care in stroke units that had proven tobeadvantageousforstrokepatientswasusedasamodelforthe discussions. 30 A large proportion of stroke patients de-velop delirium, 31 and it was expected that there would be alarge proportion of such patients included in this study. Training Concerning Caregiver-Patient Interaction Careimpliesthatacaregiverinonewayoranotherinteractswith a patient. Interaction has been described as having acontent and a relationship aspect. 32 The content aspect de-notes the activity performed, and the relationship aspectdenotes the manner in which the assistance is performedand the manner in which a caregiver provides care.In the staff education program, these concepts weredescribed and discussed, focusing on the caregivers’ inter-action with elderly patients. There was a basic assumptionthat offering the patients the opportunity to recognize andcomprehend information and to be oriented in the currentreality (knowing the time, what has happened, what ishappening, and why) would help the patients to remainoriented and decrease the risk of delirium.  Reorganization of Nursing Care The core of individualized care is to offer patients the op-portunitytousetheircapabilitiesasmuchaspossible.Intheperformance of individualized care, patients’ dependencyon caregivers may decrease, and hospitalization time maybe shortened. 22,33 The nursing care was organized according to individ-ualized care, which means that nursing activities supportthe patients individuality in the nursing care and that pa-tients perceive this care as individuals. 34–36 Small teams of one registered nurse and one licensed practical nurse wereresponsible forasmallnumberofpatientsandtheirnursingcare throughout their stay on the ward to promote conti-nuity of care and personal knowledge. The staff tried toarrange the teams’schedules so that at least one team mem-ber on each shift was familiar with the patients. There wereno differences in the proportion of staffing between the in-tervention ward and control ward. Guidance for Nursing Staff  During the 8-month study period, the nursing staff receivedguidance once a month, especially regarding problems inthe nursing care of delirious patients, from one of the au-thors(BB).Theguidancewasperformedindividuallyorinagroup and was focused on caregiver-patient interaction.Every guidance occasion took about 15 to 20minutes andstarted with the supervisor observing the nurse during anursing action and afterwards discussing how the nurse in-teracted with the patient. Control Group Care on the control ward was the usual hospital care or-ganized in a task-allocation care system; that is, the samecaregiver handled particular tasks for all patients, and nocaregiver had full responsibility for an individual patientduring his or her entire hospitalization. Several staff mem-bers could care for the patients each day, performing dif-ferent care tasks. Data Collection All patients were examined using a modified version of theOrganic Brain Syndrome (OBS) Scale, a combined obser-vation and interview scale. 37 The OBS Scale consists of twoparts: the disorientation subscale, a questionnaire contain-ing 16 items, and the confusion subscale, an observationschedule covering 39 clinical features. The confusion sub-scale was used in this study and describes various cognitive,perceptual, emotional, and personality variables; physicaland practical abilities; and fluctuations in the person’sclinical state. Instead of the disorientation subscale, theMini-Mental State Examination (MMSE) 38 was used, andinstead of the activity of daily living (ADL) subscale, theKatz ADL index 39 was used. The OBS Scale was not modi-fied in any other way.The OBS Scale has been compared with other assess-mentscalesandhasshowngoodconcurrentvalidity. 37 IthasalsobeencomparedwiththeConfusionAssessmentMethodand showed 100% agreement in a study regarding the di-agnosis of postoperative delirium in patients undergoingcoronary artery bypass graft surgery. 40 It offers a broad de-scription of reported and observed signs of organic brainsyndromes aswellasfluctuationsinapatient’s clinicalstate.The OBS Scale thus offers the opportunity to register moresymptoms and signs than a pure interview or observationinstrument. 37 One ofthreefull-timeresearchassistants (twopsychiatric licensed practical nurses and one occupationaltherapist) who had been trained to use the scale and wereexperiencedinitsusefromanotherlargestudyexaminedthepatientsusingtheOBSScaleonDays1,3,and7duringtheirstay on the ward. 41 The same research assistant usually per-formed the assessments of each patient. The research assist-ants repeatedly discussed their use of the scale to maintainconsistency. One of the three research assistants also exam-ined the patients using the MMSE and the Katz ADL indexonDays1,3,and7.Threeoftheauthorswhowereunawareof the group to which patients belonged independently di-agnosedthepatientsaccordingtoDSM-IVcriteriaregardingdelirium and dementia. In the few cases of disagreementregarding the diagnosis, each case was discussed until aconsensus was reached. These three authors collaborated inanother study using the OBS Scale; in that study, theiragreement was more than 90% in all ratings. 42 Patienthearing and vision were tested on the ward the day of ad-mission.Patientswereassessedashavingimpairedhearingif theycouldnot heara normalspeakingvoicefromadistanceof 1m with or without a hearing aid and impaired vision if they could not read the newspaper with or without glasses.Further relevant information about the patients’ livingconditions,ongoingtreatments,anddiagnoses,amongotherthings, were obtained from the patients themselves, theirspouses, nursing and medical staff, and patient records. Ethical Considerations The participating patients gave verbal informed consentafter being told about the study and that their participationor refusal would not affect their medical or nursing care. 624  LUNDSTRO ¨  M ETAL.  APRIL 2005–VOL. 53, NO. 4 JAGS  The ethical committee of the Faculty of Medicine atUmea˚ University approved the study. Statistical Analysis Univariate analyses using Student  t   test, Fisher exact test,and Pearson chi-square test were performed to describegroupdifferences,anda P o .05wasregardedasstatisticallysignificant.Tocontrolforbaselinedifferences;age,sex,andfactorsthat differed on admission between delirious patients in thetwo wards were entered into a logistic regression analysis,with delirium on Day 7 as the dependent factor. RESULTS There was no difference in the prevalence of delirium de-tected within 24 hours of admission between the interven-tion ward and the control ward (n 5 63/200, 31.5% vs 62/ 200, 31.0%; P 5 .91).Fewerpatients were delirious onDay3(123/400,30.8%vs82/400,20.5%; P o .001)thanontheday of admission, but there were no significant differencebetween the two wards (37/63, 58.7% vs 45/62, 72.6%remained delirious on Day 3;  P 5 .10), although on Day 7after admission, fewer patients remained delirious on theintervention ward than on the control ward (n 5 19/63,30.2% vs 37/62, 59.7%;  P 5 .001). These figures includedone patient in the intervention ward and three patients inthe control ward who had developed delirium after Day 1(Figure 1). One patient was assessed to be delirious on theday of discharge from the control ward during the first 7days after admission, compared with none from the inter-vention ward.Clinical characteristics of the patients in the two wardsare presented in Tables 1 and 2. Fewer delirious patients inthe intervention ward had acute stroke on admission, and alarger proportion had diabetes mellitus, but they wereequally dependent in ADLs and had the same mean MMSEscore on admission. There were no differences between pa-tients with diabetes mellitus and those with stroke in theduration of delirium. The regression analysis used to con-trol for different case mix in the two wards shows that theprevalence of delirium on Day 7 was significantly lower inthe intervention ward (Table 3). No patient with dementiaremained delirious on Day 7 in the intervention ward,compared with four patients still delirious on Day 7 in thecontrol ward.The mean length of stay  standard deviation was sig-nificantly lower on the intervention ward than on the con-trol ward (9.4  8.2 vs 13.4  12.3 days;  P o .001). Themeanlengthofstaywasalsosignificantlylowerforpatientswho were delirious on Day 1 in the intervention ward thanfor the patients who were delirious on Day 1 in the controlward (10.8  8.3 vs 20.5  17.2 days;  P o .001) (Figure 2).Despite a shorter hospital stay for patients on the in-tervention ward, there were no differences between the twowards regarding patients who could return to their ownhouse/apartment with or without home-help service (142/ 164, 86.6% vs 145/176, 82.4%;  P 5 .29). Of delirious pa-tients on the intervention ward, a larger proportion couldreturntotheirownhouse/apartmentwithorwithouthome-help service than of delirious patients on the control ward,although the difference was not significant (36/46, 78.3%vs 30/50, 60%;  P 5 .05).The delirious patients on the intervention ward hadsignificantly lower mortality during their hospital stay thanthe delirious patients on the control ward (n 5 2, 3.2% vsn 5 9, 14.5%;  P 5 .03). DISCUSSION This study on older patients admitted to general internalmedicine shows that a multidisciplinary intervention pro-gram including education, guidance, and a changed caringorganization reduces the duration of delirium, shortens thelength of the hospital stay, and reduces the mortality rateduring hospitalization for delirious patients.The intervention involved all staff on the ward, includ-ing the physicians. The reorganization of care, which im-proved the prerequisites for continuity and interactionwith patients, probably reduced the level of stress for de-lirious patients. The awareness of risk factors and precip-itating factors for delirium by all of the staff probablyresulted in improved treatment of patients with delirium.There is, however, no systematic registration of any differ-ent treatment strategies between the wards. Previous suc-cessful intervention studies in hip-fracture patients alsoinvolved all staff involved in the care of patients and in-cluded improved nursing organization and medical inter-ventions. 20–23 Interventions including only single measures,such as staff education, have not been proven effective. 43 Nor have interventions based only on consultations hadany effect if they did not involve all the staff and severalaspectsof caringforthe patients. 24,26,44 Larger proportionsof patients with dementia included in other studies mightalso have contributed to their poorer effect. 24,26,27 Toofew patients had dementia in the present study to allowanalyses of patients with dementia separately, but no pa-tient with dementia remained delirious on Day 7 in theintervention ward, compared with four patients still delir-ious on Day 7 in the control ward, which might indicatethat delirium in patients with dementia can be successfullytreated.The only earlier effective intervention study in internalmedicine patients that the authors found included severalaspects of the care and treatment of patients without de-lirium but at risk of developing delirium after admission.Theresultofthatinterventionwasareductioninnumberof days and episodes with delirium after admission, but no 200180160140120100806040200 Day 1Day 3Day 7Control wardIntervention wardDelirious patients/control wardDelirious patients/ intervention ward Figure 1.  Total number of patients and number of delirious pa-tients in the intervention and control wards, respectively, at timeof testing on Days 1, 3, and 7 after admission. MULTIFACTORIAL DELIRIUM INTERVENTION STUDY  625 JAGS APRIL 2005–VOL. 53, NO. 4  effects were seen on length of stay or mortality. 25 Thepatients in focus for the intervention in the present studywere those presenting with delirium within the first 24hours of admission, and the main outcome measures wereduration of delirium, length of hospital stay, and mortality.This makes it difficult to compare the effectiveness of thetwo programs. The authors’ interpretation is that the effectof the current intervention program is at least as effectivebecause it reduced duration of delirium, length of hospitalstay, and mortality during hospitalization for delirious pa-tients.A strength of this study was that the patients were re-peatedly assessed using cognitive testing and interviewswith staff who knew the patient best, which probablyresulted in a high detection rate for delirium. Repeatedcognitive testing seems to be necessary to detect deliriumbecause the staff did not note many cases of fluctuatingattention and cognition, which were only detected as a re-sult of the repeated and careful assessments. 45 All patientsdiagnosed as delirious or demented met the DSM-IV crite-ria, and repeated assessments using the OBS Scale and theMMSE were used for systematic assessment and documen-tation of the assessments and observations. Three assessorswithnopreviousconnectionwiththestaffonthetwowardsperformed all the testing, interviews, and registrations.Three of the coauthors, independently of each other andunaware of which group the 400 patients belonged to, as-sessed all documentation and decided whether the patientmet the DSM-IV criteria of delirium on Days 1, 3, and 7,respectively. Table 2. Diagnoses, Presenting Symptoms, MMSE Score, ADL Dependency, and Drugs for All Patients and for DeliriousPatients on the Intervention Ward and on the Control Ward on the First Day of Admission CharacteristicAll Patients Delirious PatientsIntervention Ward Control Ward P  -valueIntervention Ward Control Ward) P  -value(n 5 200) (n 5 200) (n 5 63) (n 5 62)Myocardial infarction, % 10.0 4.5 .03 7.9 0.0 .06Heart failure, % 25.5 25.5 1.00 14.3 21.0 .33Stroke, % 17.0 25.0 .05 20.6 37.1 .04Epilepsy, % 4.5 6.5 .38 9.5 6.4 .74Infection, % 19.5 15.5 .29 25.4 17.7 .30Urinary infection, % 6.5 3.5 .17 9.5 4.8 .49Fever  38 1 C, % 7.5 4.5 .21 15.9 4.0 .10Mini-Mental StateExamination score,mean  SD (n 5 193/183vs 57/50)25.1  5.8 25.3  6.2 .77 19.9  5.6 18.5  6.6 .22Proportion dependent inmore than one ADL item(n 5 199/200 vs 63/62), %47.2 50.0 .58 77.8 82.3 .53Proportion dependent in allADL items (n 5 199/200vs 63/62), %7.5 7.5 .99 14.3 17.7 .60Digitalis, % 29.0 28.0 .83 25.4 21.0 .56Diuretics, % 56.0 60.0 .42 46.0 53.2 .42Beta-blockers, % 23.0 22.5 .91 20.6 25.8 .49Calcium blockers, % 16.0 12.5 .32 12.7 14.5 .77Insulin, % 19.0 11.0 .03 15.9 8.1 .18Analgesics, % 31.0 26.5 .32 28.6 21.0 .33Benzodiazepines, % 6.5 7.0 .84 4.8 8.1 .49Neuroleptics, % 10.0 9.0 .73 17.5 17.7 .97Number of drugs,mean  SD4.0  2.9 3.9  2.7 .67 3.5  2.6 3.6  2.6 .93 ADL 5 activity of daily living according to Katz ADL-index; SD 5 standard deviation; MMSE 5 Mini-Mental State Examination. Table 3. Logistic Regression Analysis to Control for Base-line Differences Between the Delirious Patients in the TwoWards (N 5 125) Independent Variable Odds Ratio 95% Confidence IntervalWard 3.12 1.43–6.81Stroke on admission 1.44 0.62–3.35Sex 1.35 0.59–3.05Age 1.01 0.95–1.08Diabetes mellitus 0.53 0.22–1.27 Note : Included variables: ward, age, sex, and those variables that differed be-tween the wards in patients with delirium on admission.Model chi-square 15.233;  P 5 .009. 626  LUNDSTRO ¨  M ETAL.  APRIL 2005–VOL. 53, NO. 4 JAGS
Search
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks