Critical Appraisal Skills Program (CASP) randomized control trial checklist
According to Moule and Goodman (2009), critical appraisal is the process of assessing a study or research project to determine its quality and value in informing best practice. Critical appraisal of research is an essential principle in evidence-based practice. Assessments should be objective and based on a structured approach that uses recognised criteria. This critique will use a combination of criteria suggested by Coughlan, et al. (2007) and the Critical Appraisal Skills Program (CASP) randomized control trial checklist.
11.0 Credibility
The authors of each of the three studies in this critique are highly qualified in their specialties and have exceptional levels of knowledge of the subject, making the results credible. Eva Simonse, the lead author of the first article, is an experienced medical doctor working in the Department of Paediatrics in a Hospital in the Netherlands. She has written other journal articles on the breastmilk. Her co-authors are PhD holders in related fields. The authors of the second article all work in the Department of Paediatrics or Neonatology in their respective university hospitals in Spain. The authors of the third article are all PhD holders working in paediatric departments in hospitals in Brazil and Canada. Furthermore, the findings in all three articles are consistent despite the samples coming from Europe and Australia.
12.0 Purpose of the Study
The clarity of purpose in a study plays a vital role in helping to assess its significance. Further, if the question posed as the subject of research is too broad, the results may lack applicability, so the narrower and more focused the research, the more useful the results (Smith, 2009). All three studies examined in this paper expressed the purpose of their research clearly. In STUDY 1, the purpose of the study was clearly captured in the objective and justified in the background section. In STUDY 2, the researchers justified the study as important due to the limited research evidence available on the use of expressed breast milk (EBM) in providing pain relief to preterm infants. In STUDY 3, the purpose of the study was clearly stated in the objective.
Sampling
Randomization in 1 of the 3 groups in STUDY 1 was done by an independent employee using the method of sequentially numbered and opaque sealed envelopes. The sample size was 75 infants. The sample in the study was sufficiently described together with the exclusion criteria. Experienced laboratory employees did blood sampling in a standardized manner. The study accounted for all the participants.
In Study 2, the sample population was described, and the exclusion criteria provided. A single dose was given to the two groups receiving either EMB or sucrose. The study did not include a control group. The sample size was 63 infants. Randomization was done electronically after probabilistic sampling was done. All parents and researchers in the study except for the investigator who administered treatment were blinded, eliminating the possibility of researcher bias. The study accounted for all the participants.
The research design of STUDY 3 was a non-inferior controlled trial using a sample of 113 participants. As in STUDY 2, there was no control group; participants received either EBM of 25% glucose. Randomization was performed using the Statistical Analysis System version 8.2.
The three studies accounted for all participants. The study accounted for all participants from the beginning to the end of the study.
Methods
All studies chose pain levels measured against PIPP criteria as the primary outcome. The Premature infant Pain Profile (PPIP) scale used in the studies is a validated and reliable method of assessing and scoring infant pain.
In STUDY 1, the procedure was recorded in on video. Three investigators were trained before the study in assessing the PPIP and COMFORTneo scales. In STUDY 1, patients and investigators were not blinded for the allocated intervention. The investigators analysed the videos, but since they were not blinded for the allocated treatment, it suggests bias. In the same randomization was not by an independent employee, the analysis of the videos would also have involved independent investors in minimizing bias. Pain induced by heel lance in premature neonates evaluated with the PPIP scale was the primary outcome. The secondary outcome of the study was the pain assessed by the COMFORTneo scale. Statistical methods for the quantitative and qualitative portions of the data were described. The descriptive statistics included 95% confidence intervals or means with standard deviations.
Like STUDY 1, the primary outcome was pain assessed using the PPIP. The secondary outcomes of the study included sociodemographic variables, duration of crying, incidence, and percentage of crying, and the clinical variables of Apgar score and mode of delivery.
Like the other two studies, the primary outcome in Study three was pain assessed using the PPIP. Secondary outcomes were incidence of crying, duration of crying, and adverse events. Statistical analysis was performed using Minitab 15.1.
Ethics
The researchers in STUDY 1 sought approval of the study protocol and informed consent forms from the regional and local medical ethics committee. In STUDY 2, the parents provided written informed consent, and the Hospital Universitario Gregorio Maranon Ethics Committee approved the study, which was also registered in the ClinicalTrials.gov database. The local research ethics committee approved the study protocol, and the study was registered at the Australian and New Zealand Clinical Trials Registry.
Results
The results were presented in both tables and narration. Analysis of covariance was an appropriate method of data analysis. Additionally, the reliability of judging the videos of pain assessment was analysed. The use of a crossover trial made the results of STUDY 2 more generalizable than those of STUDY 1 and STUDY 3. In STUDY 3, the results of the study were presented in tables, figures, and narration. The results were presented in a table and sufficiently narrated.
Limitations
The strength of STUDY 1 was recording of pain procedures, which allows future referencing.
In the study by citation 1, the main limitation was the impossibility to blind the intervention, introducing bias. Another limitation was the mean age of at the time of heal lance that was close to 37 weeks. The mean age affects the generalization of the results to extremely preterm infants.
The discussion section was comprehensive and cohesively tied information in the background and results together. The conclusion was on point, reflecting the results and the discussion.
The strength of STUDY 2 was its crossover design, which allowed for the adjustment of results for potential subjective variables. The limitations of the study included the uncontrollable confounders that could have influenced the pain intensity, pain procedures were not recorded, and it was not possible to calculate interobserver variability. However, the investigators attended a joint training session in a move to standardize their PPIP scores. STUDY 3 fails to discuss its limitations. Failure to indicate the limitations of the study, such as the use of a sample from a singular location, negatively affects the generalization of the results.