Introduction
Methods
Eligibility criteria
Information sources
Search strategy
Study selection
Data collection process
Data items
• Item |
• First authora |
• Year of publicationa |
• Countrya |
• Study designa |
• Length of studya |
• Study populationa |
• Inclusion and exclusion criteriaa |
• Definitions of adverse drug events/adverse drug reactions/iatrogenic disease/iatrogenic eventa |
• Definition and/or criteria for causality assessment methoda |
• Definition of severitya |
• Definition of preventabilitya |
• Definition of predictabilitya |
• Statistical analysesa |
• Population characteristicsb |
• Incidence of adverse drug eventsb |
• Drugs impliedb |
• Causality resultsb |
• Items required for causality assessment methodsb |
• Severity aspectsb |
• Predictabilityb |
• Clinical features of adverse drug eventsb |
• Length of stay in the ICUb |
• Costs of hospitalisation in the ICUb |
Risk of bias assessment
Summary measures
Synthesis of results
Additional analysis
Quality assessment
Item | Recommendation | Points granted |
---|---|---|
Title
| ||
Explicit topic | One must understand that the article concerns ADE that require ICU admissions | 2 |
Abstract
| ||
Type(s) of ICU studied | Describe the type of ICU (for example, medical, surgical) | 2 |
Causality assessment method(s) | Indicate the causality assessment method(s) used | 2 |
Proportion of patients admitted to the ICU for ADE | The denominator should be the total number of patients admitted in the studied ICU during the observation period | 2 |
Preventability rate of ADE | Indicate the estimated preventability rate of ADE | 1 |
Study duration | Indicate the study observation period(s) | 1 |
Introduction
| ||
Background | Explain the scientific background and rationale for the investigation being reported | 2 |
Objectives | State the objectives of the study | 2 |
Methods
| ||
Mandatory
| ||
Description of the study design | A prospective patient screening is preferred in order to avoid missing data | 3 |
Type(s) of ICU studied | Describe the type of ICU(s) (for example, medical, surgical) | 3 |
Complementary information on the setting environment | Indicate the presence of eventual other ICU(s) in the hospital, and mention specific wards (oncology, haematology, geriatrics). This information may help in appraising and understanding results | 3 |
Description of the study size rationale | Study size should be argued | 3 |
Definition of ADE | The definition of the institute of medicine and recommended by Nebeker and colleagues [4] should be preferred | 3 |
Evaluation of inter-rater reliability for inclusion decision | Indicate how inter-rater reliability for inclusion decision was assessed | 3 |
Description of evaluators’ training | Describe the profession of evaluators and, if applicable, participation in specific training for the study | 3 |
Description of patients’ screening | Describe who was in charge of the patients’ screening and how screening was performed | 3 |
Description of inclusion/exclusion criteria | Describe and justify inclusion and exclusion criteria | 3 |
Description of collected data and outcomes measured | Collected data should include characteristics of study participants (age, gender, severity score at admission (SOFA/SAPS II), number and classes of drug(s) involved; see items in Results) | 3 |
Description of drug history collecting method | Describe the sources of data used for establishing drug history, including all patients’ prescriptions (home, hospital). If possible, the patient or a relative should be questioned to identify all drugs prescribed, all drugs taken in self-medication and drugs prescribed but not taken (inobservance) during the month prior to ICU admission. If patients were already hospitalised before ICU admission, all drugs administered during the hospital stay should be collected | 3 |
Description of causality assessment method(s) | Mention the causality assessment method(s) used. Assessment of inter-rater reliability would be welcome | 3 |
Description of preventability method/criteria | Mention the criteria used for assessing preventability. Assessment of inter-rater reliability would be welcome | 3 |
Definition of the severity | Mention the severity of the ADE: fatal (ADE contributed to death), life-threatening (ADE requiring organ supply) and moderate (ADE only requiring monitoring) | 3 |
Study duration | Mention the date of beginning and ending of the study | 3 |
Not mandatory
| ||
Description of statistical analysis (if applicable) | Describe all statistical methods, if applicable | 0.5 |
Research of medical causes that contributed to ADE (for preventable drug events) | Indicate how the medical causes that contributed to ADE were investigated (that is, drug interactions, contraindications between drugs and patient’s disease, nonappropriate dosage) | 0.5 |
Results
| ||
Mandatory
| ||
Proportion of patients admitted to the ICU for ADE according to the chosen denominator | The chosen denominator should be the total number of included patients admitted to the ICU during the study observation period | 3 |
Results for inter-rater reliability for inclusion decision | Describe the analysis results of inter-rater agreements/disagreements | 3 |
Description of the characteristics of patients with ADE | Describe the studied population: age, gender, severity score at admission, reason for admission, origin of patients (home, hospital) | 3 |
Number and classes of drugs suspected to be involved in the ADE responsible for ICU admission | The Anatomical Therapeutic Chemical classification should be used | 3 |
Results for causality assessment | Provide all causality assessment results | 3 |
Results for the preventability rate of ADE | Indicate the estimated preventability rate of ADE | 3 |
Results for severity of ADE | Indicate how many patients died and how many required organ support | 3 |
Results for ICU mortality rate of patients with and without ADE (separately) | Indicate and compare the ICU mortality rates of patients with and without ADE. Estimates of the hospital mortality rates for these patients would also be welcome | 3 |
Length of stay in ICU of patients with and without ADE (separately) | Indicate and compare the lengths of stay in the ICU for patients with and without ADE | 3 |
Not mandatory
| ||
Results of inter-rater reliability for causality and preventability | Describe the analysis results of inter-rater agreements/disagreements | 0.5 |
Research of medical causes that contributed to ADE (for preventable ADE) | For preventable ADE, investigations into the medical causes that contributed to ADE, such as prescription despite contraindication, dosage nonappropriate according to weight or specific pathologies (that is, renal impairment) may constitute valuable data to report | 0.5 |
Number of drugs taken by patients with ADE prior to ICU admission | The total number of drugs taken by patients with ADE prior to ICU admission would be welcome, as well as drugs prescribed but not taken | 0.25 |
Clinical features of ADE | Describe all clinical features of ADE | 0.25 |
Comorbidities of patients with ADE | Describe comorbidities of patients with ADE | 0.25 |
Discussion
| ||
Limitations | Discuss limitations of the study, taking into account sources of potential bias | 2 |
Interpretation | Interpret results and compare with previous studies | 2 |
Sensitivity analysis
Results
Study selection
Study characteristics
Study | Country | Study duration | Study design | Type of hospital | Type of ICU | Number of patients admitted during studied period | Aims of the study |
---|---|---|---|---|---|---|---|
Trunet and colleagues, 1986 [10] | France | 33 months (August 1978 to April 1981) | Prospective monocentre | Teaching hospital | Multidisciplinary ICU | 1,651 | Determine cause and effect relationship between drugs and adverse event, severity of DII, role of underlying disease and potential preventability of DII |
IGICE, 1987 [11] | Italy | 6 months | Prospective (data collection on a given day each week) | ND | 27 general ICUs | 4,537 | Document aspects of ADR epidemiology in 27 general ICUs |
Nelson and Talbert, 1996 [12] | USA | 1 month (July to August 1993) | Prospective monocentre | Teaching hospital | Medical ICU, CCU and internal medicine service | 127a | Describe the frequency and pattern of drug-related morbidity that results in hospital admission and the extent to which these admissions are avoidable |
Darchy and colleagues, 1999 [13] | France | 12 months (January to December 1994) | Retrospective monocentre | General hospital (500 beds) | Medico-surgical ICU (15 beds), CCU (6 beds) | 623 | Determine whether aging of the general population and medical advances have altered the incidence, causes and consequences of severe IDs, compared with Trunet and colleagues first study [21] |
Hammerman and Kapeliovich, 2000 [14] | Israel | 36 months (July 1994 to June 1997) | Prospective monocentre | Teaching hospital (900 beds) | CCU (9 beds) | 2,559 | Evaluate major cardiac iatrogenic disease as the cause of admission to the CCU |
Lehmann and colleagues, 2005 [15] | USA | 12 months (November 1998 to November 1999) | Prospective monocentre | Four teaching hospitals | Four surgical ICUs, three medical ICUs and one paediatric ICU | 5,727 | Identify the frequency and type of iatrogenic medical events requiring admission to the ICU. Assess the consequences of iatrogenic medical events for patients, and the incidence of disclosure of iatrogenic medical events to patients |
Grenouillet-Delacre and colleagues, 2007 [16] | France | 6 months (May to October 2003) | Prospective monocentre | Teaching hospital | Medical ICU | 436 | Assess the characteristics of life-threatening ADR in patients admitted to a medical ICU in order to identify associated risk factors that could facilitate early identification |
Rivkin, 2007 [17] | USA | 19 weeks (December 2004 to May 2005) | Prospective monocentre | Teaching hospital (1,076 beds) | Medical ICU (12 beds) | 281 | Determine frequency, severity and preventability of ADR leading to admission to a medical ICU |
Schwake and colleagues, 2009 [18] | Germany | 12 months (January to December 2003) | Prospective monocentre | Teaching hospital (1,685 beds) | Medical ICU (14 beds) | 1,554 | Determine the incidence of ICU admissions due to ADR and compare affected patients with patients admitted to the ICU for the treatment of deliberate self-poisoning using medical drugs |
Mercier and colleagues, 2010 [19] | France | 6 months (November 1999 to April 2000) | Prospective monocentre | Teaching hospital | Medical ICU (27 beds) | 528 | Determine the incidence, risk factors, severity and preventability of IEs as cause of ICU admission |
Nazer and colleagues, 2013 [20] | Jordan | 5 months (August to December 2010) | Prospective monocentre | Teaching cancer centre (170 beds) | Medico-surgical ICU (12 beds) | 249 | Describe the incidence, characteristics and cost of ADE that necessitate admission to the ICU in oncology patients |
Risk of bias within studies
Study | |||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|
Evaluated item | [10] | [11] | [12] | [13] | [14] | [15] | [16] | [17] | [18] | [19] | [20] |
Prospective study design? | 1 | 1 | 1 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
Description of the type of studied ICU? | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 0 | 1 |
Complementary information on the setting environment? | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 1 | 1 | 0 | 0 |
Study size rationale? | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
Definition of ADE according to IOM? | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 |
Evaluation of inter-rater reliability for inclusion decision? | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
Description of evaluators’ training? | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
Description of patients’ screening? | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
Description of inclusion/exclusion criteria? | 1 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
Description of collected data? | 1 | 1 | 0 | 1 | 0 | 1 | 1 | 0 | 1 | 1 | 1 |
Description of the drug history collecting method? | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 |
Description of causality assessment method? | 1 | 0 | 1 | 1 | 1 | 0 | 1 | 1 | 1 | 1 | 0 |
Description of preventability method/criteria? | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 0 | 1 | 1 |
Definition of ADE severity? | 1 | 1 | 0 | 1 | 0 | 0 | 1 | 1 | 1 | 0 | 0 |
Description of study duration? | 1 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
Results for incidence of ADE requiring ICU admission? | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
Results for inter-rater reliability for inclusion decision ? | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
Description of the characteristics of patients with ADE (age, gender, severity score at admission, reason of admission, origin of patients)? | 0 | 1 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 |
Description of number and classes of drugs suspected to be involved in the ADE responsible for ICU admission? | 1 | 0 | 0 | 1 | 1 | 0 | 1 | 0 | 1 | 1 | 1 |
Results for causality assessment? | 1 | 0 | 0 | 1 | 0 | 0 | 1 | 1 | 0 | 0 | 0 |
Results for preventability rate? | 0 | 0 | 0 | 1 | 1 | 0 | 1 | 1 | 0 | 1 | 1 |
Results for ADE severity? | 1 | 1 | 0 | 1 | 0 | 0 | 1 | 1 | 1 | 0 | 1 |
Results for ICU mortality rate of patients with ADE? | 1 | 1 | 0 | 1 | 0 | 0 | 1 | 1 | 1 | 0 | 1 |
Results for the length of stay for patients with and without ADE (separately)? | 0 | 0 | 0 | 1 | 0 | 0 | 1 | 1 | 1 | 0 | 1 |
Proportion of items completely reported | 0.58 | 0.42 | 0.33 | 0.71 | 0.46 | 0.38 | 0.79 | 0.67 | 0.63 | 0.46 | 0.63 |
Results of individual studies
Population characteristics
Clarification of terminology and incidence of adverse drug events
Characterisation of adverse drug events responsible for admissions to the ICU
Study | Severity/type of scale and results, % ( n ) | Drugs involved a, % of patients (n) | Preventability and categorisation if any | Preventability rate, % of patients ( n ) |
---|---|---|---|---|
Trunet and colleagues, 1986 [10] | Fatal, 9.3% (9); life-threatening, 27.9% (27); severe, 35% (34); moderate, 27.8% (27) | Psychotropic drugs, 17.5% (17); anticoagulants, 13.4% (13); intravenous solutions, 12.4% (12); antibiotics, 11.4% (11); diuretics, 9.3% (9) | Not investigated | Not investigated |
IGICE, 1987 [11] | Minor, 0%; major, 100% | 30 drugs involved in ADR; single drug, 83.3% ADR (20); association of drugs, 16.7% ADR (4) | Not investigated | Not investigated |
Nelson and Talbert, 1996 [12] | Not investigated | ND specifically for ICU patients: hypoglycaemic drugs, 15.8% (12)b; diuretics, 13.2% (10)b; anti-infectious drugs, 11.8% (9)b; cardiovascular drugs, 10.5% (8)b; psychotic drugs, 9.2% (7)b | Definitely avoidable – satisfied by one of the following criteria: patient did not take a prescribed drug, known allergy to the drug, contraindication between the drug and his disease/condition, the patient took a drug not prescribed or not indicated for a diagnosed disease | ND specifically for ICU patients: definitely avoidable, 49.3% (36)b |
Possibly avoidable: monitoring of the patient’s drug therapy not inadequate | Possibly avoidable, 9.6%b (7) | |||
Not avoidable – no reasonable actions could have prevented it | Not avoidable, 37.0%b (27) | |||
Unevaluable – information is insufficient to make a determination or is contradictory | Unevaluable, 4.1%b (3) | |||
Darchy and colleagues, 1999 [13] | Fatal, 14.6% (6); life-threatening, 12.2% (5); moderate, 73.2% (30) | 55 drugs involved in ID; single drug, 22 IDs; association of drugs, 19 IDs; diuretics, 17.1% (7); oral anticoagulants, 14.6% (6); nonsteroidal anti-inflammatory, 14.6% (6); antibiotics, 14.6% (6); anaesthesia, 12.2% (5) | Event that should not occur if management is the best that medical science can provide | 73.1% (30) |
Hammerman and Kapeliovich, 2000 [14] | Not investigated | 234 drugs involved in major IE; single drug, one major IE; association of drugs, 63 major IEs; nitrates, 76.6% (49); diuretics, 70.3% (45); beta-blockers, 68.8% (44); ACE inhibitors, 45.3% (29); calcium antagonists, 43.8% (28) | Event that could have been avoided if the prescription of therapy had respected the art of medical practice | 64.1% (41) |
Lehmann and colleagues, 2005 [15] | Not investigated | Narcotic analgesics, 42.8% (9); sedative hypnotics, 23.8% (5) | Event avoidable using any means currently available, unless those means where not considered standard of care [22] | ND specifically for ICU patients: preventable, 34.4% (22); not preventable, 14.1% (9); not assessable, 51.6% (33) |
Grenouillet-Delacre and colleagues, 2007 [16] | According to World Health Organization [23]: life-threatening, 94% (124); potentially life-threatening, 6.0% (8); among these events, 15.9% (21) contributed to death | 132 drugs involved; psychotropic drugs, 22.5% (25); immunosuppressive drugs, 21.6% (24); anticoagulant drugs, 13.5% (15); anti-infectious drugs, 12.6% (14); antihypertensive drugs, 12.6% (14) | Definitely preventable – all conditions for avoidance of its occurrence were fulfilled | Definitely preventable, 10.6% (14) |
Potentially preventable – not all conditions were met to avoid its occurrence | Potentially preventable, 37.1% (49) | |||
Not preventable – treatment procedure consistent with current knowledge of good medical practice | Not preventable, 21.2% (28) | |||
Not assessable (lack of data), 31.1% (41) | ||||
Rivkin, 2007 [17] | Fatal, 19% (4); severe, 66.7% (14); moderate, 14.3% (3) | 39 drugs involved; single drug, 43% ADR (9); associations of drugs, 57% ADR (12) | Medication use was inappropriate and contrary to standard clinical practice [24] | 85.7% (18) |
Schwake and colleagues, 2009 [18] | Life-threatening, 37.4% (37); potentially life-threatening, 62.6% (62); Among these events 2% (2) were fatal | anticoagulants, 62.6% (62); analgesics, 25.2% (25); diuretics, 16.2% (16); antihypertensives, 5% (5); antidepressants, 5% (5) | Not investigated | Not investigated |
Mercier and colleagues, 2010 [19] | Not investigated | 54 drugs involved; chemotherapy, immunosuppressant drugs, 27.8% (15); psychotropic drugs, 14.8% (8); cardiovascular drugs, 14.8% (8); anaesthesia, analgesic drugs, 11.1% (6); oral anticoagulants, 9.3% (5) | Preventable ADR: drug not used according to the summary of product characteristic | 64% (32) |
Nazer and colleagues, 2012 [20] | Fatal, 28.1% (16); life-threatening, 17.6% (10); significant, 54.3% (31) | Antineoplastic drugs: 64.9% (37); analgesics: 15.8% (9); anticoagulants: 7% (4); others: 12.3% (7) | ADE met at least one of the following criteria: inappropriate drug or unnecessary for the patient’s condition, drug dose; route or frequency inappropriate for the patient’s age, weight or disease state; required supportive/preventive therapies not prescribed; required therapeutic drug monitoring or laboratory tests not performed; history of allergy; resulted from a well-established drug interaction | 17.5% (10) |
Causality results
Required data or study | Causality assessment methods | |||||
---|---|---|---|---|---|---|
Kramer and colleagues [25] | WHO [26] | Naranjo and colleagues [27] | Karch-Lasagna [28] | Hallas and colleagues [29] | Begaud and colleagues [30] | |
Required data for causality assessment | L, Chron, D, R, LT, AEC | L, Chron, D, R, AEC, PP | L, Chron, D, R, Pl, DM, LT, AEC, Atcd, OE | L, Chron, D, R, AEC | Chron, D, R, LT, AET, Atcd | L, Chron, D, R, LT, AEC, PP |
Trunet and colleagues, 1986 [10] | de, 29.7% (30); pr, 45.6% (46); po, 20.8% (21); un, 3.9% (4) | |||||
IGICE, 1987 [11]a | ||||||
Nelson and Talbert, 1996 [12] | pr, 59.6% (31)b; po, 40.4% (21)b | de, 15.3% (8)b; pr, 40.4% (21)b; po, 25% (13)b; un, 19.3% (10)b | ||||
Darchy and colleagues, 1999 [13] | de, 34.1% (14); pr, 34.1% (14); po, 31.8% (13) | |||||
Hammerman and Kapeliovich, 2000 [14] | ND | |||||
Lehmann and colleagues, 2005 [15] | ND | |||||
Grenouillet-Delacre and colleagues, 2007 [16] | ND | Very likely, 8.3% (11); likely, 51.5% (68); possible, 40.2% (53) | ||||
Rivkin, 2007 [17] | de, 4.8% (1); pr, 80.9% (17); po, 14.3% (3) | |||||
Schwake and colleagues, 2009 [18] | ND | |||||
Mercier and colleagues, 2010 [19] | ND | |||||
Nazer and colleagues, 2013 [20]a |