Systematic and critical review of research studies. Prof. dr Tiny Jaarsma

Transcription

1 Systematic and critical review of research studies Prof. dr Tiny Jaarsma

2 One day: - You will use reports/articles of others - Are they good, where are gaps and flaws? One day: - You will be a reviewer for an article /grant /doctoral seminar /dissertation - Is this scientifically sound? One day: - Your report/article will be reviewed

3

4 Why? What is the question? Is it relevant? Will answering it contribute new knowledge to the field? How do I answer these? Your own expertise The literature is your friend Seek advice from mentors/colleagues

5 How? Can the team answer the question? Track record We have previously shown Collaborators Infrastructure/equipment

6 Overall quality Scientific quality Others: Implementation Usability Suitability (e.g for particular funds ) Use of finances

7

8 Plaisance, 2004 Poorly written Grammar, spelling, typographic errors Disorganized Weak content Inaccurate information Lack of clarity Format does not fit the Journal Poor title Jargon unfamiliar to readers Not conform priorities

9

10

11 Language (style, errors and typo s)

12 The language error categories were divided into three principal groups: grammatical, structural and lexical Coates et al., 2002

13

14

15 Language: grammar and syntax Language: style and structure

16 Language (style, errors and typo s) Sloppiness Balance in text: heading/subheading References Tables and figures Balance Placement Colors

17 A picture tells a thousand words Powerful selling points Explanation of paradigms, overall picture

18 myocardial oxygen demand angina pectoris ST-segment depression elevation of capillary wedge pressure perfusion defect regional systolic dysfunction regional diastolic dysfunction normal LV function exercise time + load

19

20 Example bad table

21

22 Overall quality Scientific quality Others: Implementation Usability Suitability (e.g for particular funds ) Use of finances

23 CONSORT(Consolidated Standards of Reporting Trials) CONSORT extended STROBE: STrengthening the Reporting of OBservational studies in Epidemiology. M.A.A.R.I.E. framework TREND: Transparent Reporting of Evaluations with Nonrandomized Designs COREQ: Consolidated criteria for reporting qualitative research

24

25 25

26

27

28

29

30

31

32

33

34

35 The CONSORT 2010 Statement thus addresses the minimum criteria, although that should not deter authors from including other information if they consider it important. Schulz, Ann Intern Med. 2010

36 36

37 37

38 38

39 8: Assignment Method Method used to assign units to study conditions, including details of any restriction (e.g., blocking, stratification, minimization) Inclusion of aspects employed to help minimize potential bias induced due to non-randomization (e.g., matching) 15: Baseline equivalence Data on study group equivalence at baseline and statistical methods used to control for baseline differences 39

40 1a. Identification as a randomized trial in the title 1b. Structured summary of trial design, methods, results, and conclusions

41 41

42 42

43 Abstract (256 words) Ghrelin causes interdigestive contractions of the stomach in rats. However, it remains unknown whether ghrelin causes interdigestive contractions in the small intestine. Four strain gauge transducers were implanted on the antrum, duodenum, proximal and distal jejunum. After an overnight fast, gastrointestinal (GI) contractions were recorded in freely moving conscious rats. Spontaneous phase Ⅲ-like contractions were observed at every min in rat GI tract. The fasted motor patterns were replaced by the fed motor pattern immediately after food intake. Two minutes after finishing the spontaneous phase Ⅲ-like contractions in the antrum, acyl ghrelin (0.8, 2.4 and 8.0 mg/kg per min) was continuously infused for 30 min. Three-five minutes after the starting ghrelin infusion, augmented phase Ⅲ-like contractions were observed at the antrum, duodenum, and jejunum. Ghrelin infusion (0.8, 2.4 and 8.0 mg/kg per min) significantly increased motility index of phase Ⅲ-like contractions at the antrum and jejunum in a dose dependent manner, compared to that of saline injection. Thus, it is likely that exogenously administered ghrelin causes phase Ⅲ-like contraction at the antrum, which migrates to the duodenum and jejunum. The possible role of 5-HT, in addition to ghrelin, in mediating intestinal migrating motor complex (MMC), is discussed The WJG Press. All rights reserved 43

44 44

45 The concepts in the title are not clear Abbreviation in the title The title is not covering the subject Wrong headings The abstract has references Secondary outcomes in the abstract The conclusion is outside this study 45

46 46

47 2a 2b Scientific background and explanation of rationale Specific objectives or hypotheses

48 Why is this study done? What is known? Where are the gaps 48

49 49

50 Not clear what this study adds (what is the problem?) Mix introduction and discussion To much problems described: will this study solve all these problems? To much references/to little references 50

51

52

53 3a 3b Description of trial design (such as parallel, factorial) including allocation ratio Important changes to methods after trial commencement (such as eligibility criteria), with reasons

54 54

55 55

56 Not complete Intervention study Redundant Prospective quantitative randomized controlled trial with a follow-up 56

57 4a Eligibility criteria for participants 4b Settings and locations where the data were collected

58

59 Eligibility criteria are not reported Not clear how patients are recruited (referral, self selection) Setting is not clear (primary, secondary, or tertiary health care or from the community?) Location is not clear How many centers were involved?

60 The interventions for each group with sufficient details to allow replication, including how and when they were actually administered

61 The study included 2 types of intervention and a control treatment. Figure 1 shows the intensity of contacts with the cardiologist and nurse specializing in management of patients with HF (HF nurse) during follow-up. All patients received usual routine management by their cardiologist. This included an outpatient visit less than 2 months after hospital discharge and then every 6 months. Because 17 hospitals were involved, with potential differences in followup intensity, a standardized follow-up schedule was developed. Patients in the 2 support groups were visited by an HF nurse during admission, for education and support, and were scheduled for additional visits at the outpatient clinic. 61

62 6a 6b Completely defined pre-specified primary and secondary outcome measures, including how and when they were assessed Any changes to trial outcomes after the trial commenced, with reasons

63

64 7a 7b How sample size was determined When applicable, explanation of any interim analyses and stopping guidelines

65 8a 8b Method used to generate the random allocation sequence Type of randomization; details of any restriction (such as blocking and block size) Mechanism used to implement the random allocation sequence (such as sequentially numbered containers),

66 If done, who was blinded after assignment to interventions (for example, participants, care providers, those assessing outcomes) and how

67 Blinding of key trial person participants, health care providers, or other caregivers that rely on the category of treatment (surgical/technical procedure, participative interventions, devices) and comparator assessed Blinding of outcome assessors depending on the primary outcome considered Blinding of statistician

68 The ferric carboxymaltose solution is dark brown in appearance and is therefore easily distinguishable from the placebo (0.9% saline). Hence, unblinded study personnel (at least one physician) not involved in any study assessments are responsible for preparing and administering the study treatment injections in black syringes and using a curtain (or similar) to maintain patient blinding. Unblinded personnel make sure the patient is not able to observe the preparation of study treatment injections. To ensure the blinded investigator remains blinded, post-baseline iron values are sent by the central laboratory only to the unblinded personnel, who then are responsible for evaluating the presence of elevated iron levels or severe anaemia. In case of elevated iron parameters or severe anaemia, the unblinded physician amends the study treatment as described above.

69 Main inclusion criteria: NYHA class II / III, LVEF 40% (NYHA II) or 45% (NYHA III) Hb: g/dL Iron deficiency: serum ferritin <100 µg/l or <300 µg/l, if TSAT <20% Treatment adjustment algorithm: Interruption: Hb>16.0g/dL or ferritin>800µg/l or ferritin>500µg/l, if TSAT>50% Restart: Hb <16.0g/dL and serum ferritin <400µg/L and TSAT<45% Blinding: Clinical staff: unblinded and blinded personnel Patients: usage of curtains and black syringes for injections *total dose required for repletion calculated using the Ganzoni formula Anker et al, Eur J Heart Failure 2009;11:

70 70

71 12a Statistical methods used to compare groups for primary and secondary outcomes 12b Methods for additional analyses, such as subgroup analyses and adjusted analyses

72 Statistics/Analysis Are all statistical tests described Are the statistical tests relevant? Are statistical assumptions fulfilled, for example independency, normality, linearity, etc? Is the sample size relevant for the statistical tests? Results/Findings Are the findings clearly described? Is the CONSORT diagram (flow in the trial) described in case of trials? Are tables and figures properly described; are there too many? 72

73 13. Participant flow 14. Recruitment 15. Baseline data 16. Numbers analysed 17. Outcomes and estimation 18. Ancillary analyses 19. Harm

74

75

76 2957 patients screened with Symptomatic Heart Failure Excluded 1626 No informed consent patients illegible randomised 348 Control 348 Basic A&C 353 Intensive A&C Died during admission 9 Control 8 Basic A&C 9 Intensive A&C analysed 339 Control 340 Basic A&C 344 Intensive A&C

77 336 Enrolled in HF management program or other trial 328 Dementia, a psychiatric diagnoses or deafness 292 PCI, HTX, or valve replacement was planned or had recently taken place 125 End-stage of another terminal illness 419 Logistic problems (too short admitted to be recruited, or no transportation to the hospital available)

78 On average only 26% (range 5-67%) of initially screened patients was included in the trials (n=27 trials) (Gonseth 2004)

79 Screened In included Rich months 280 (22%) Ekman months 158 (9%) Stewart (3%) Strömberg months 106 (5%) DEAL-HF months 240 (30%) COACH monhts 1049 (36%)

80 For each primary and secondary outcome, a summary of results for each estimation study condition, and the estimated effect size and a confidence interval to indicate the precision 80

81 81

82 Incomplete report on patient flow Patients are missing To many analysis New analysis not related to the aim Unclear reporting Imbalance between tables and figures 82

83

84 Limitations Addressing sources of potential bias, imprecision Generalizability External validity, applicability Interpretation Consistent with results, balancing benefits and harms, and considering other relevant evidence

85 85

86 To long All findings are discussed Repetition of the results New analysis /results in the discussion To little to many limitations To many implications, not related to the results 86

87 87