I was wondering how much it would be if I got some statical analysis help for my dissertation. I am doing my analysis on the R programme. Basically I’ve done half of my analysis in terms of descriptives but I’m not sure if the script is right. Please let me know ASAP as my paper is due within a few weeks and I am really struggling with the lack of support given to me by my university.
Method Plan GLAD The data used for this study was from participants who were involved in the Genetic Links to Anxiety and Depression (GLAD) Study, launched in 2018 by the National Institute for Health Research (Davies et al., 2019). The main aim for the GLAD study was to collect genetic and phenotypic data from 40,000 participants with lifetime anxiety and/or depression and understand where genetic and lifestyle links may coincide with one another. The GLAD study still recruits new participants, however for the purpose of this study only participants who have completed the GLAD study before the 2nd of February 2020 will be used (n =44974). Consent for the GLAD study was conducted online using the following website: (https://gladstudy.org.uk/). Participants were required using several channels including advertising on social media, clinical recruitment via the National Health Service (NHS) as well as word of mouth. (Davies et al., 2019). The GLAD study is conducted in to two parts, the sign-up questionnaire which asks participants for demographic data and other personal information. The second part includes a DNA sample kit and an additional questionnaire. However. for the use of this study only the information obtained from the initial sign-up questionnaire will be used (Davies et al., 2019). COPING The Coronavirus Outbreak Psychological Experiences (COPING) study is another branch of the GLAD study whereby GLAD participants were asked to complete the PHQ9 (Patient Health Questionnaire) during various stages of the COVID-19 pandemic. The details of the PHQ9 questionnaire will be outlined below. All the COPING surveys were completed during the COVID-19 pandemic. The first (baseline) survey was released on 19th May 2020 which is then compared with the PHQ9 results at GLAD sign-up to identify the effects of the pandemic. The COPING study still recruits new participants, however for the purpose of this study only participants who have completed the COPING study before June 1st 2020 will be used (n =13812). Consent for the COPING study was conducted online using the following website: (https://kclbs.eu.qualtrics.com/jfe/form/SV_5hHYPH37qvK0Cyh) . Participants were required using several channels including advertising on social media, clinical recruitment via the National Health Service (NHS) as well as word of mouth. (Breen et al., 2020) The COPING study is conducted in only one part, with two questionnaires. The first questionnaire is the sign-up questionnaire which asks participants for demographic data and other personal information. The second questionnaire includes the PHQ9 questionnaire, which is completed through various timepoints throughout the pandemic. However. for the use of this study only the information obtained from the initial sign-up questionnaire and the questionnaire obtained during the first nationwide lockdown on June 1st. Questionnaire information PHQ9 The Patient Health Questionnaire is a depression questionnaire that can be used monitor the severity of an individual’s depression severity and how they are currently responding to treatment. This questionnaire however is not a screening tool for depression, but the questions are based on the 9 DSM-IV criteria on a scale from “0” (not at all) to “3” (nearly every day). The maximum score that can be achieved by this question is 27, a score between 20-27 indicates an extremely high depression severity. A score of 15-19 indicated a moderately severe depression severity, 10-14 is described as moderate, 5-9 as mild and 0-4 as no depression severity. This questionnaire was conducted twice for all participants, (once at the GLAD sign-up and once at the COPING sign-up). (https://patient.info/doctor/patient-health-questionnaire-phq-9) LIFE The LIFE questionnaire is a set of questions based on the participants social life including the nature of the participant’s relationship, friendships and other family members. The questionnaire asks whether they are feeling supported by these groups of people and if they are valued including questions about their behaviour towards the participants. This questionnaire also asks about whether they have been in any dangerous situations, suffered any forms of abuse and what they do for fun. The LIFE questionnaire has 28 questions are scored either by a likehart scale or through nominal methods. This questionnaire was conducted once for all participants, during the GLAD sign-up. EPDS (Edinburgh Postnatal Depression Scale) The Edinburgh Postnatal Depression Scale is a set of screening questions that can help indicate whether a parent is experiencing symptoms of postpartum mental illness including depression and anxiety. However, this questionnaire is not an official diagnosis. The EPDS contains 10 short statements which are scored from a scale from “0” (not at all) to “3” (nearly every day). The maximum score that can be achieved by this question is 30, a score above 12-13 indicates that there is a possible risk that the parent is suffering from a postpartum mental illness. This questionnaire was conducted once for all participants, during the GLAD sign-up. (http://www.perinatalservicesbc.ca/health-professionals/professional-resources/health-promo/edinburgh-postnatal-depression-scale-(epds)) MHD (Mental Health Disorder) The Mental Health Questionnaire is a set of questions that ask the participant whether they have been officially diagnosed with a mental health disorder by a professional and which disorder they currently have. The questionnaire is not a screening tool for mental illness, but it does ask to what extent the illness effects their daily life. The MHD contains 53 questions asking participants their mental health history which were scored either “Yes,” No”, “Don’t Know” or “Prefer not to Answer”. Participants were able to state one of those answers to say many disorders as they wanted, as well as writing their own disorder in a text box and the end of the questionnaire. This questionnaire was conducted once for all participants, during the GLAD sign-up. All questionnaire data were acquired using Qualtrics survey software (Qualtrics, Provo, UT). Questionnaires were asked to be completed at GLAD sign-up and were only required to me filled out once. The PHQ9 was the only questionnaire that was asked to be filled out twice, once at GLAD sing-up and once at COPING sign-up. The EPDS questionnaire was only required to be completd by women who have either recorded experiencing PPD in their MHD questionnaire or in their initial GLAD sign-up questionnaires. GLAD and COPING Study Participants For participants from the GLAD and COPING studies to be included in this study, they must be female UK residents, over the age of 18, and meet the MHD questionnaire criteria for either postpartum depression (PPD) or major depressive disorder (MDD). For the GLAD participants they must have completed their initial sign-up before the 2nd of February whereas the COPING participants must have completed their sign-ups before the 1st of June. Further analysis of the demographics of these two groups will be explained further in the results section. Edited Common Subject Assessment The Edited Common Subject Assessment, developed by NIHR BioResource Centre Maudsley, is included in the GLAD Study sign-up questionnaire. This provided socio-demographic information for the present study, such as participants’ age and sex. Ethical approval Ethical approval was granted by: i) NHS Health Research Authority, South West- Central Bristol Research Ethics Committee (20/SW/0078; COPING), and ii) Psychiatry, Nursing and Midwifery Research Ethics Committee at King’s College London (HR-19/20-18157; RAMP). Information sheets, consent forms and questionnaires were reviewed by the Feasibility and Acceptability Support Team for Researchers and the Service User Advisory Group. Data Cleaning information The data for this project will be analyzed using R and will involve using several steps. The first step would be to clean all the data sets and questionnaire data. This process would involve removing participants based on missing or incomplete data, converting all text or categorical variables to numeric variables and exclude implausible values by setting them to NA. The data cleaning would also ensure that all data meets the stated inclusion criteria for this study as well as meeting the date deadlines required for the GLAD and Coping participants. The following data sets would be cleaned: · PHQ9_GLAD (Current depression scores at GLAD sign-up) · MHD_GLAD (Mental Health Diagnosis at GLAD sign-up) · LIFE_GLAD (Information on marital status and support at GLAD sign-up) · DEM_GLAD (Demographic information for GLAD surveys) · PHQ9_COPING_GLAD (Depression scored during the pandemic) · EPDS_GLAD (Postpartum scores for the GLAD dataset) · DEM_COPING_GLAD (Demographic information for pandemic surveys) Descriptive analysis The next step involves creating an experimental/case sample (women who have experienced PDD) and a control sample (women who have not experienced PPD but have experienced MDD) based on inclusions and exclusion criteria. The inclusion criteria for the control sample will include: 1. Women 2. Have reported receiving a diagnosis of depression in the MHD questionnaire 3. Report being over the age of 18 4. Completed the GLAD demographics, MHD, PHQ9, LIFE survey and the baseline COPING surveys The inclusion criteria for the case sample will include: 1. Women 2. Must have had at least one child 3. Have reported receiving a diagnosis of postpartum depression in the MHD questionnaire 4. Report being over the age of 18 5. Completed the GLAD demographics, PHQ9, LIFE and EPDS surveys and the baseline COPING survey Here I will also conduct statistical tests to compare the groups for any significant differences in the following categories: T-test · Age · PHQ9 score (at GLAD sign-up and the first COPING survey) Chi-Square tests · Ethnicity (white vs BAME) · Marital status (living with a partner vs not living with partner/no partner) · Employment (unemployed vs employed) · Qualifications (A-Levels/ Level 3 qualifications vs No A-Levels/level 3 qualifications) Main Analysis The last step will involve performing regression analyses on the dataset using the R programming language to look at risk factors for worsening depression during the COVID-19 pandemic in the two groups. To do this, I will need to create a binary variable (worsening depression vs non-worsening depression) in the postpartum women group (case group) and the general women group (control group). PHQ9 scores that have been increased by 3 or more points will be referred to as a clinical increase. The proportion of women whose PHQ9 score has increased by at least 3 points will be reported for both groups in the descriptive statistics. The women who have higher PHQ9 scores during the COVID-19 pandemic compared to their GLAD sign-up survey will be categorized as "worsening depression". For women who have the same or lower PHQ9 scores during the COVID-19 pandemic compared to the GLAD sign-up study, they will be categorized as "non-worsening depression" I will do this for both samples of women (postpartum depression vs general depression) I will perform logistic regressions, regressing the worsening/non-worsening depression variables onto the following risk factors: · Ethnicity (white vs POC) · Marital status (Relationship cohabiting/ Relationship not cohabiting) · Employment (unemployed vs employed) · Qualifications (A-Levels/ Level 3 qualifications vs No A-Levels/level 3 qualifications) · Relationship score (High/healthy vs low/not healthy) · Number of previous children (continuous variable) · Living with children in the pandemic vs not living with children in the pandemic · Living with children before the pandemic vs not living with children in the pandemic · Comorbid mental disorders (any comorbid disorders/ no comorbid disorders) I will perform these regressions individually to assess their beta coefficients (i.e. I will perform each of these regressions for the case and the control sample). I will then combine these into one multivariable regression model. Davies, M. R. et al. The Genetic Links