Lecture 1: Evidence Base and Critical Evaluation PDF

**[Lecture 1: Evidence Base and Critical Evaluation]** +-----------------------------------+-----------------------------------+ | **Question** | **Answers** | +===================================+===================================+ | 1. **What is the pu...

**[Lecture 1: Evidence Base and Critical Evaluation]** +-----------------------------------+-----------------------------------+ | **Question** | **Answers** | +===================================+===================================+ | 1. **What is the purpose of | The purpose of research evidence | | research evidence and what | is to provide appropriate | | sort of questions would be of | assessment and treatments to | | interest?** | patients. Assessments in | | | psychological settings are | | | investigations into the | | | difficulties an individual is | | | suffering from as well as using | | | research to compare thoughts, | | | feelings, behaviours and symptoms | | | with groups that share similar | | | and different characteristics | | | that the individual is | | | experiencing. An assessment will | | | often require the use of | | | diagnostic criteria, for example | | | using the DSM-5 or ICD 11 as a | | | manual tool to identify specific | | | psychological disorders. | | | Furthermore, an assessment raises | | | the awareness of heterogeneity of | | | disorders as the same disorder | | | can manifest itself differently | | | in everyone. Regarding | | | treatments, these are therapeutic | | | interventions that are aimed at | | | alleviating the symptoms of a | | | disorder that a person may be | | | experiencing. These can either be | | | psychological or biological in | | | nature such as to treat | | | depression clinicians may advise | | | patient to participate in CBT | | | and/or take SSRI's. By providing | | | treatments that require research | | | evidence, clinicians can monitor | | | the progress of a treatment and | | | change in the patient. | | | | | | When using research evidence, the | | | most important questions that are | | | of interest to a clinician are: | | | | | | - **Comorbidity** of disorders | | | as they can overlap, for | | | example depression can also | | | imbricate with anxiety or | | | substance abuse disorders. | | | Therefore, it can manifest as | | | one of the symptoms within a | | | different disorder. | | | | | | - **Influencing factors of | | | culture** and context | | | (gender) on a disorder, for | | | example with depression | | | within society is a stigma | | | and therefore individuals | | | lack support and acceptance | | | of the disorder (Sue & | | | Morishima, 1982; Ng, 1997). | | | (Zhang et al., 1998) found | | | that only 3% of Asians | | | (living in the US) would seek | | | help with depression as | | | society would perceive a | | | mental health disorder as a | | | poor lineage within an | | | individual\'s family. | | | | | | - **Prognosis** of a disorder; | | | understanding the progress | | | and change of a disorder over | | | time. For example, with | | | depression, specifically | | | major depressive disorder, if | | | untreated can last for 6-12 | | | months, whereby 2/3 will have | | | suicide thoughts **(Bains & | | | Abdijadid, 2023).** | | | | | | - **Gaps in treatment;** within | | | the NHS there are long | | | waiting times, therefore GP's | | | physicians would prescribe | | | SSRI's and TCAs to treat | | | depression, then incorporate | | | other techniques such as CBT. | | | Therefore, for a prominent | | | disorder such as depression, | | | clinicians would be | | | interested in understanding | | | whether these techniques are | | | effective as there is an | | | array of treatments and | | | interventions. | | | | | | - **Prevalence** of a disorder: | | | for example, the prevalence | | | of depression is 12% over an | | | person's lifetime Therefore, | | | highlighting the commonality | | | of depression in life. | | | | | | - **Aetiology:** the cause of a | | | disorder, it has been | | | reported that Major | | | depressive disorder is | | | believed to be | | | multifactorial, including | | | biological, genetic, | | | environmental, and | | | psychosocial factors (Bains & | | | Abdijadid, 2023). | | | | | | - **Diagnosis**: currently the | | | diagnosis of depression is | | | based off the DSM-5 or | | | ICD-11, however currently | | | machine learning has been | | | implemented to provide a more | | | detailed and disorder focused | | | analysis of the | | | diagnosis/assessments and | | | methodologies used to | | | diagnosis an array of | | | disorders including | | | depression from 33 articles. | | | It suggests that using | | | machine learning to | | | understand how disorders are | | | diagnosed can be used a | | | predictor for other disorders | | | (Iyortsuun et al, 2023). | | | | | | - **Genetic predisposition**: | | | provides evidence on how | | | genes can manifest disorders, | | | for example depression has | | | found to be a predisposition | | | of the serotonin gene in the | | | neurochemical mechanisms | | | within the brain, therefore | | | leading to mood disturbances | | | (Wurtman, 2005). | | | | | | - **Treatment engagement:** is | | | an important component in | | | practice as attendance to | | | sessions provides better | | | results to a treatment, for | | | example major depressive | | | disorder (Kwan et al, 2010). | +-----------------------------------+-----------------------------------+ | 2. **What is the importance of | EBP: evidence-based practice is | | EBP and how does this relate | the integration of t**he best | | to the 'Three-Legged | available research with clinical | | Stool'?** | expertise in the context of | | | patient characteristics, culture | | | and preferences.** (APA, 2005). | | | The purpose of EBP is to promote | | | effective psychological practice | | | and enhance public health by | | | applying empirically supported | | | principles of psychological | | | assessment, case formulation, | | | therapeutic relationship and | | | interventions. (APA, 2005). | | | | | | EBP is important as it has been | | | found to improve patient outcomes | | | and healthcare systems which | | | benefit public health care and | | | professionalism within health | | | care setting (Connor et al, | | | 2023). | | | | | | The 'Three-Legged Stool' is a | | | framework for how EBP is used to | | | make effective clinical decisions | | | within practice; it was | | | conceptualized by Sackett et al. | | | in 1996. | | | | | | The three legs within this | | | framework refer to the following: | | | | | | 1. **Clinical Expertise:** | | | | | | The expertise of a clinician | | | refers to the way in which a | | | clinician communicates to a | | | patient, assesses a patient\'s | | | disorder, creates a rapport or | | | alliance with the patient and how | | | they deliver treatments and most | | | importantly providing empirically | | | supported treatments. | | | Unfortunately, there is no | | | quantifiable measure of how | | | expertise of a clinician is | | | implemented within practice as it | | | is subject to the delivery of the | | | skills and knowledge a clinician | | | accumulates over their | | | profession. It has been noted | | | that clinical expertise is an art | | | form with its complex in nature | | | (Peterson et al, 2016). Thus, | | | highlighting the difficulty of | | | measuring this component but also | | | its necessity in practice. | | | | | | 2. **Patients\' values, | | | characteristics, preferences | | | and circumstances:** | | | | | | This component emphasizes the | | | importance of understanding the | | | alternatives, risks & benefits, | | | preferences and access of | | | treatment and assessments for the | | | patient. | | | | | | Within the NHS, a patient does | | | not have freedom to choose | | | treatment or assessment as the | | | NHS generally allocate the | | | appropriate treatment based of | | | the most prevalent disorder being | | | presents, in contrast private | | | healthcare systems give choices | | | and alternative to their patients | | | as they are paying for the | | | service, this includes making | | | changes when necessary and | | | discussing/monitoring treatment | | | interventions. | | | | | | This disparity is not just the | | | cost of treatment but also the | | | accessibility of treatment, under | | | public healthcare the waiting | | | times will exacerbate the | | | disorder further, whereas private | | | healthcare will allow a patient | | | to address a disorder | | | immediately, hence emphasizing | | | the importance of time and cost | | | within practice (Owusu‐Frimpong | | | et al., 2010). This can also be | | | applied to patients who live in | | | third world countries that have | | | no accessibility to appropriate | | | healthcare such as in | | | | | | Overall, the patients\' | | | circumstances play a role within | | | clinical decision-making. | | | Furthermore, cultural influences | | | and preferences affect treatment | | | and assessment as for example | | | depression is seen as a stigma | | | and therefore would need to be | | | addressed within practice. | | | | | | 3. **Best available research | | | evidence:** | | | | | | There are three components to | | | this factor: | | | | | | 1. Synthesizer | | | | | | The process of finding and | | | amalgamating literature, this | | | includes reading meta-analysis or | | | systematic-analyses and | | | critically evaluating the | | | appraisal of publications. | | | | | | 2. Evidence User | | | | | | This involves extracting the | | | information that is reliably | | | relevant to practice which can | | | consequently be integrated. | | | | | | 3. Researcher | | | | | | This is the individual of groups | | | of people that design, conduct, | | | analyse and report their study | | | findings which are then used by | | | other researchers, clinicians, | | | doctors, etc. | | | | | | The type of literature or | | | research that is used can vary | | | from reviewing existing research, | | | articles or continually | | | evaluating current methods within | | | practice. Furthermore, clinicians | | | may opt to conduct their own | | | research study in order to | | | establish or witness the efficacy | | | of new or current treatment, | | | diagnosis or interventions. | | | However, it is dependent on the | | | research question the researcher | | | has designed their study upon. | | | Different studies will have | | | different research aims and | | | methods, therefore clinicians | | | will use the best research that | | | is closely related to their way | | | of practice. For example, a | | | psychiatrist would use papers on | | | the latest neurochemical | | | pharmaceutical interventions to | | | treat major depressive disorder, | | | whereas as a psychologist or a | | | therapist may want to understand | | | the most effective treatment for | | | depression using CBT or | | | mindfulness. | | | | | | These are important as it is | | | hypothesis driven, in that | | | practitioners will examine which | | | research best answers their | | | questions of interest, | | | furthermore, using empirically | | | supported evidence ensures them | | | that the interventions or | | | information, such as on etiology | | | or comorbidity, is valid. More | | | so, it allows clinicians to | | | establish the best treatment | | | approaches through rigorous | | | analysis of current research | | | available (APA, 2006). | | | | | | Allows clinicians to adopt new | | | and current methods for | | | intervention. For example, | | | incorporating mindfulness as a | | | tool to maintain and alleviate | | | the symptoms of depression has | | | been shown to be an effective | | | technique (Szymonik & Szopa, | | | 2024). This also fosters a | | | research mindset in clinical | | | practice and requires clinicians | | | to keep up with current research. | +-----------------------------------+-----------------------------------+ | 4. **What is the hierarchy of | The hierarchy of evidence is a | | evidence, explaining each | pyramid framework that assumes | | tier?** | that other sources of evidence | | | are better than others. | | | | | | At the bottom we have | | | **individual case studies**: | | | | | | These are based on single cases | | | that are very difficult to | | | generalize. They are useful when | | | examining a case study with | | | in-depth complex detail. For | | | example, Phineas Gage in 1848 | | | helped understand how damage to | | | the corpus callosum can result in | | | a personality change. However, | | | these lack ecological validity. | | | Furthermore, it is risky to use | | | an individual case as an approach | | | for treatment, furthermore, is | | | difficult to quantity how this | | | individual case can be | | | generalised to other patients. | | | | | | Next is **qualitative research**: | | | these are in-depth case studies | | | of small groups of people, | | | essentially an individual case | | | study on a large group. These are | | | subjective lived experiences that | | | generate rich data. It can be | | | used to identify themes which can | | | be used across multiple | | | individuals undertaking the same | | | treatment or experiences. | | | | | | **Cohort studies**: | | | | | | Are longitudinal observational | | | studies on large groups of people | | | with an aim to investigate the | | | outcome of a specific treatment | | | for example, they typically | | | investigate the epidemiology of a | | | certain health outcome. It is | | | used to measure the success of | | | failure of a treatment which is | | | compared in two different groups. | | | | | | These can be classified as | | | prospective or retrospective and | | | mixed. Prospective study is the | | | superior subtype as it looks at | | | the outcomes in the future, | | | whereas retrospective examines | | | the outcomes in the past. This is | | | because prospective studies can | | | control for variables, examine | | | the outcomes and predictors (Wang | | | & Katan, 2020). | | | | | | Prospective studies are accurate | | | on the data collection as they | | | are taken over time, whereas | | | retrospective use pre-existing | | | data, which is time and cost | | | effective, however this may lead | | | to information biases. | | | | | | The advantages of using cohort | | | studies are that a researcher can | | | apply multiple outcomes with a | | | single predictor. For example: | | | examining childhood adversity and | | | depression, researchers can | | | design a study that looks at | | | childhood adversity of neglect | | | and abuse to depression in later | | | life. Furthermore, cohort studies | | | allow greater statistical power | | | due to the large sample size | | | needed, and as they are | | | longitudinal in nature they | | | provide temporal insight into how | | | the outcomes and predictors are | | | related to one another. However, | | | overall cohort studies are | | | expensive to run, and they are | | | subject to attrition (drop --out) | | | and they cannot control for all | | | confounds such as in the example | | | above researcher would not be | | | able to control for substance | | | abuse, genetics, environmental | | | influence, SES and etc. | | | Furthermore, results from cohort | | | studies do not reflect the | | | population, lack ecological | | | validity, especially when no | | | inclusion and exclusion criteria | | | is included. Additionally, in a | | | cohort we can not assume that | | | correlation means causation as | | | there is no control group | | | incorporated. | | | | | | RCT: randomized control trials | | | are a gold standard within | | | clinical research (Littlewood, | | | 2011). These are very effective | | | in investigating the causal | | | associations within research, for | | | example looking at the | | | effectiveness of interventions | | | for depression on generalised | | | depressive disorder. In a RCT | | | participants would be allocated | | | to either intervention group | | | \[receive CBT\] or control \[no | | | CBT\]. RCT have a protocol called | | | the CONSORT (consolidated | | | standard for reporting trials), | | | these are guidelines that state | | | how to run RCT, these include | | | scientific background, RCT label, | | | report eligibility criteria, | | | funding. RCT are very | | | advantageous as allow researchers | | | to draw causal inferences about | | | interventions, they are easy to | | | replicate, less subjective to | | | interpretation and useful when | | | comparing individual patients | | | with similar symptomology. | | | Saturni et al (2014) highlighted | | | that RCTs offer great insight | | | into the efficacy of data under | | | control conditions and provide | | | treatment effectiveness in | | | diverse population, thus | | | highlighting that RCT's control | | | for confounds. Especially in | | | double blind studies, as it | | | reduces biases within research. | | | | | | Overall, RCT's seem effective | | | however they lack external | | | validity, in that it can not be | | | used to compare treatment causes | | | and lacks to explain the | | | relationship between the group | | | allocation and treatments. These | | | are more group focused than | | | patient and thus lack in-depth | | | analysis. Finally RCT are | | | difficult to conduct as the | | | eligibility is stringent, | | | including time-consuming and | | | expensive. | | | | | | Meta -- analysis are reports that | | | merge single reports into one | | | whole publication. These are | | | reader friendly as they reduce | | | the burden on a reader, reduce | | | noise in literature and control | | | confounds and they provide better | | | statistical power. They provide | | | an overall navigational finding | | | for clinicians to implement | | | treatments or interventions | | | within practice. The aim of a | | | meta-analysis is to synthesize | | | findings from multiple studies. | | | They combine results from many | | | studies by using a statistical | | | approach which is useful for | | | resolving inconsistencies across | | | studies and increase power for a | | | more precise estimate of effect | | | size. These minimize biases while | | | maximizing data, it assess the | | | heterogeneity within studies. | | | There are three way in which | | | researchers can determine the | | | heterogeneity within literature: | | | | | | 1. Chi-square analysis; a test | | | that measure the model of | | | data to the observed data, | | | this can result to misleading | | | data as the test has a low | | | power to detect heterogeneity | | | in small sample size | | | | | | 2. Visual inspection of data | | | using forest plot or | | | meta-graph; these are visual | | | presentation of the overlaps | | | within studies which would | | | indicate heterogeneity. | | | | | | 3. *I*^2^ statistic; this is the | | | most common method; it | | | calculates the percentage of | | | variability that is explained | | | by the heterogeneity across | | | studies. | | | | | | At the of the tier is systematic | | | studies, these are studies that | | | integrate all studies into one | | | report, however they can either | | | be qualitative or quantitative | | | and include very specific | | | inclusion criteria. These are | | | effective resources as they | | | amalgamate many studies into one | | | with stringent inclusion | | | criteria, which in turn reduce | | | potential biases and subjective | | | interpretations. Systematic | | | studies use predetermined terms | | | when collecting literature, these | | | are relevant to the research | | | question in interest and of high | | | quality (Jaded et al, 1996). | | | Systematic studies are relatively | | | unbiased due to predetermined | | | selection, the synthesis of | | | studies draws an overall | | | conclusion, less subjective, | | | increases power relative to | | | individuals\' studies and reduces | | | literature burden readers. | | | | | | The limitations of these studies | | | are that there may be a | | | publication bias, as these | | | studies are costly and | | | time-consuming industries may | | | fund researchers therefore, they | | | are pressured to find significant | | | findings. Synthesizing studies is | | | difficult as individual studies | | | vary in their own respective | | | methodologies; this also includes | | | data extraction. Furthermore, the | | | inclusion and exclusion criteria | | | in studies can affect the volume | | | of data for extraction and | | | consequently can affect the | | | quality of data as it will lack | | | in detail. | | | | | | Good research would incorporate | | | control groups, control for | | | confounds, random sampling, valid | | | and reliable measures, consider | | | ecological validity and clinical | | | utility. Control groups are very | | | important as they provide a | | | baseline for studies to draw | | | inferences. Sample size effects | | | and significant findings if a | | | study shows a significant finding | | | it is meaningless unless | | | considering the effect size and | | | confidence intervals as spurious | | | results can occur which is | | | important in replicating studies. | | | | | | In general, within research there | | | is a lack of longitudinal studies | | | these are important as they | | | provide detail about the progress | | | of treatment approaches, | | | likelihood of triggers or | | | relapse, long-term side effects | | | of treatments, risk and disorder | | | development and long-term | | | prognosis. These are difficult | | | because they are expensive to run | | | and time consuming, also drop-out | | | is a common effect in these | | | studies and there is a wide | | | heterogeneity of a disorder | | | across individuals as experiences | | | differ over time. | | | | | | To deal with these limitations | | | there are protocols in place: | | | | | | NICE guidelines provide an | | | interpretation of the best | | | research which is unbiased for | | | clinicians to use for research | | | and practical purposes. They | | | indicate the trade-off between | | | high quality research and | | | applicability to practice. | | | | | | Open Science: reduces publication | | | bias and replication crisis as it | | | requires researchers to provide | | | their study hypothesis, | | | methodology and aims before data | | | collection to keep them | | | accountable on the outcomes of | | | their research. | +-----------------------------------+-----------------------------------+