The Canadian Depression Research and Intervention Network (CDRIN) Maritimes Registry: Reliable Assessment to Enable Research and Improve Clinical Care
The Canadian Depression Research and Intervention Network (CDRIN) Maritimes Registry: Reliable Assessment to Enable Research and Improve Clinical Care
While effective interventions for depression exist, their success rates are unsatisfactory and their provision is haphazard. The Canadian Depression Research and Intervention Network (CDRIN) Maritimes Depression Hub will improve the delivery of care and the quality of outcomes for youths, adults and seniors with depression across the Maritimes. The investigators will establish an integrated system of assessment, treatment, research and education related to depression with the active involvement of those with lived experience. The establishment of a patient registry is a key step that will facilitate evaluation and reform of current services, integration of patient choice and community resources into treatment programs, monitoring long-term outcomes, and development of more effective treatment approaches through research. The registry will facilitate research that will include validation of new diagnostic and outcome measurement tools, low-cost clinical trials and collaborative projects with national and international partners. Educational programs will involve training the next generation of researchers, those with lived experience, clinicians, and health system managers in critical appraisal and will facilitate their involvement in research. The registry, the proposed systematic measurement of outcomes and the broad dissemination of information and skills will improve the quality of research and of care as well as the experience of patients and their families.
The need for a registry: It is increasingly recognized that major advances in the treatment of mental disorders will require large scale clinical research. Recently demonstrated ways of completing large-scale research with finite resources include the routine use of electronic health records (EHR), data linkage and randomized registry trial. Use of EHR is the most efficient way of rapidly obtaining large amounts of information. However, EHR cannot completely exclude confounding by indication and other unmeasured variables. Therefore, tests of treatment effects require experimental designs that cannot be replaced by routine health records data. The gold standard for testing the effects of treatment in an unbiased way is the randomized controlled trial (RCT), where measured and unmeasured confounders are balanced through the randomization process and any remaining confounding is due to chance alone. RCTs are valued as the highest level of evidence, but are costly and take significant time to be completed, partly because of the need to screen a large group of individuals to identify eligible participants.
The most efficient unbiased test of interventions, new treatment modalities and novel ways of treatment delivery is a method that combines EHR use with the randomized controlled trial (RCT) methodology: the randomized registry trial (RRT). The RRT takes advantage of a registry of individuals with available information to identify a large number of individuals suitable for an RCT. The RRT approach is efficient especially if the same information (e.g. diagnosis and treatment history) is used repeatedly for different purposes. The same information can be used for clinical purposes, service improvement and multiple research projects. RRT will allow obtaining answers about the efficacy of new treatments and management strategies significantly faster and at a much lower cost than traditional RCTs. Therefore, the investigators propose to establish a registry that has the capacity to conduct RRTs. The proposed registry will be integrated with similar efforts across Canada. Jointly, this collaborative network of registries will facilitate fast and economical testing of new treatments, which is urgently needed to advance the therapeutic options for people with depression and related conditions.
Registry The CDRIN Maritimes registry will provide an integrated platform for clinical research and service evaluation. Participation in the registry will be offered to individuals seeking help for mental disorders across primary, secondary and tertiary care settings in Nova Scotia. In the spirit of participatory research, all participants in the registry will become members of the registry and will be kept informed of registry development, as well as educational and training opportunities. Those who have consented to be contacted regarding future research opportunities will be invited to actively participate in research.
Number of participants The number of participants in the registry is not set in advance and will depend on funding and needs of service development and research studies that will use the registry. The registry will be scalable and the investigators expect to enrol between 1000 and 30,000 participants. For the additional reliability study, numbers of participant have been determined based on a power calculation. A power calculation suggests that between 313 and 400 individuals assessed twice are required to establish the reliability of diagnostic instruments with an acceptably small measurement error (confidence interval for reliability coefficient alpha of 0.15 or smaller for disorders present in 10% of participants or more). Therefore, the investigators propose to recruit 400 individuals for the reliability study.
Inclusion and exclusion criteria The registry will be broadly inclusive with minimal exclusion criteria. Any individual who is seeking or receiving help for mental health related problems in Nova Scotia is eligible to become a member of the registry. Residents of Nova Scotia will be eligible to enter the Registry if they understand English or French and have a capacity to provide informed consent or if an appropriate substitute decision maker provides consent on their behalf. There are no exclusion criteria related to age, sex, or ethnicity. Whilst the recruitment will be targeted to include individuals suffering with depression and other types of mental illness, no specific diagnosis is required for participation in the registry. Recruitment is diagnosis-independent as the goal of the registry is to create an inclusive database of individuals seeking mental health care in the Maritimes. As such, no specific formal diagnosis is necessary. Many individuals will present at various primary, secondary, and tertiary health care settings who will not meet full diagnostic criteria for a mental health condition, but could still be sub threshold and meet diagnostic criteria at a later time. It is important to include these individuals in the registry to facilitate the clinical care of those seeking mental health care and also to facilitate research on various subsets of individuals seeking mental health care, with and without formal mental health diagnoses. Furthermore, restricting participation to people who only fulfill diagnostic criteria would not allow for testing of the reliability of the diagnosis or the boundaries between disorders and normality. Recruitment will be targeted toward individuals suffering from a potential diagnosis of Depression as it is one of the most common and burdensome mental health conditions in Canada. Participants will not be excluded from the registry if they are found to not meet diagnostic criteria following diagnostic interviews.
Participant enrollment Potential participants will be approached at the point of registration, in waiting rooms and during clinical consultations in primary, secondary and tertiary care services that are likely to serve people suffering from depression and other types of mental illness. A clinician or researcher will offer participation in the registry. After explanation of the registry purpose and the need for reliability studies, participants will provide written informed consent for their information to be kept in a registry and used for research on the causation, classification and treatment of depression and other mental disorders.
The participants will also be given the option to participate in instrument reliability studies that may include repeated completion of the same instruments.
Assessments The registry will include a minimal set of core assessments (Tables 1 and 2) and a broader palette of optional instruments (Tables 3 and 4) for screening, diagnosis and current state (symptom profile, severity and functioning) assessment for youth (Table 1 and 3) and adult (Tables 2 and 4) participants. The selection of assessment instruments beyond the minimal set will be determined in conjunction with participant, clinician and investigator preferences. It is not expected that any participant would complete all optional measures at any given time. Selection of instruments will take into account burden to participants, so that no single assessment takes more than 3 hours. Tables 1 and 2 list the minimal sets of instruments for youth and adult participants. Table 3 shows the broad palette of optional instruments for youth and table 4 shows the full palette of assessment instruments for adults. Screening and status assessment instruments will be administered repeatedly (typically once every 3-6 months, but more often if needed to assess an effect of recently initiated treatment or change in care delivery). Diagnostic instruments will typically be administered once (twice for participants in a reliability study), but may be repeated in after 1 to 5 years to assess changes in psychopathology profile and diagnoses.
Screening instruments Strengths and Difficulties Questionnaire (SDQ) will be the core screening instrument for youth participants up to age 16. SDQ is a 25-item measure of psychopathology for children and adolescents with parent-report and self-report versions validated for ages 3-16 and 11-16 respectively. Simple and accessible language makes SDQ easy to complete. Balance of positive and negative items makes this measure highly acceptable. The total difficulties scale / emotional subscale is a highly sensitive screener for depression and anxiety disorders in youth. The questionnaire takes only 5 minutes to complete and detects depression and other disorders with an accuracy equal to or higher than longer instruments. An established cut-off value of 14 on the total difficulties scale of SDQ will be used to select participants who are likely to suffer from depression or related disorders.
Patient Health Questionnaire (PHQ-9) will be the core screening instrument for adults and older adolescents (from age 17). PHQ-9 is a 9-item self-report questionnaire that has been extensively validated for use with adults and, more recently, adolescents. PHQ-9 is highly sensitive in screening for depression, with a score of 10 or higher suggestive of diagnosis. Thanks to its brevity and straightforward response format, PHQ-9 is easy and quick to complete, has been found to be acceptable and is widely used in primary and secondary care practice.
Mood and Feelings Questionnaire (MFQ) will be an alternative screening measure for older children and adolescents that is more directly focussed on depression. MFQ has two versions: a long 33-item version and a short 13-item version. The investigators will use the short MFQ, which has been validated for 7- to 18-year old participants and has been found to be acceptable with a cut-off of 11 or more being highly sensitive in detecting a diagnosis of major depressive disorder.
Screen for Child Anxiety Related Emotional Disorders (SCARED) and the Generalized Anxiety Disorder 7-item scale (GAD-7) will be used as optional screening measures for services and investigators interested in detecting anxiety disorders in addition to depression. SCARED self-report and parent-report versions have been validated for children and adolescents aged 8-18, with a high-sensitivity threshold of 20 for detecting anxiety disorders. The self-report and parent-report versions have been found to be psychometrically equivalent. GAD-7 has been validated as a highly sensitive screen for generalized anxiety disorder in adults. GAD-7 is moderately specific for generalized anxiety and detects a range of anxiety disorders.
Everyday Feelings Questionnaire (EFQ) will be optionally used to screen for depression and anxiety disorders among parents of child and adolescent participants. EFQ is a brief 10-item measure of current depressive and anxiety symptoms with a balance of positive and negative items making it highly acceptable, non- stigmatizing and suitable for a brief assessment of emotional wellbeing in the parents of children and adolescents presenting to health services. Availability of a spouse-report version facilitates obtaining information on both parents when only one parent is available for direct assessment. EFQ has excellent psychometric properties and strong convergent validity with established measures of depression and anxiety.
Diagnostic instruments Kiddie Schedule for Affective Disorders and Schizophrenia - Present and Lifetime version, adapted for the Diagnostic and Statistical Manual of Mental Disorders (DSM)-5th Edition (K-SADS -5). K-SADS -5 is a semi-structured diagnostic interview to establish current and lifetime diagnoses of mental disorders in children and adolescents based on interviews with parent and with the young person. Previous versions of K-SADS have been validated for children and adolescents aged 7-18 and have been found to have high inter-rater reliability. K-SADS has recently been adapted to DSM-5 by Dr Joan Kaufman and colleagues at Yale University (as K-SADS-5) and is awaiting a test of reliability. Based on the K-SADS-5, the investigators will derive best estimate DSM-5 diagnoses established in consensus meetings involving experienced psychiatrists blind to the diagnoses of parents. Information provided by parents will be submitted for use in these meetings after being checked to be free of any indication of parental diagnosis. Our team has received training in using K-SADS by the team of John Nurnberger and will be receiving additional training on the DSM-5 adaptation with Dr Joan Kaufmann in September 2014.
Structured Clinical Interview for DSM-5 (SCID-5) is a semi-structured diagnostic interview for adults designed to be administered by a clinically trained interviewer. While earlier versions of SCID have been shown to be highly acceptable and reliable, the recently developed SCID-5 is awaiting its test of reliability. SCID-5 provides exact wording of questions to ask for specific diagnostic symptoms but allows additional probing to establish the symptom presence and timing with adequate accuracy and make a judgment about its clinical significance. SCID-5 provides algorithmic rules that guide the interviewer through DSM-5 diagnoses and allows skipping questions that will not make a difference to a diagnostic conclusion, given the information already provided. In October 2013 Dr. Michael First (Columbia University) trained the team of investigators at Dalhousie in using SCID-5. Based on our preliminary experience, a SCID-5 interview can last between 45 minutes and 3 hours, depending on the complexity of psychopathology, efficiency of the interviewer and communication skills of the interviewee. The interviewers will be clinically trained nurses, social workers, occupational therapists, psychologists or psychiatrists. They will initially receive a three-day training from Dr. First, the author of SCID-5 and a collaborator in this project. Following this, the investigators will conduct once monthly meetings to review recorded interviews, discuss intervening issues and maintain the quality of interview administration. The investigators expect most of the interviews (70%) to be carried out by research coordinators (nurses/social workers/occupational therapists) and a smaller proportion (30%) to be carried out by the investigators and clinicians (psychiatrists and clinical psychologists). The investigators will record the qualifications and experience level of each interviewer.
Status assessment: symptom severity and functioning Current severity of depressive, anxiety and psychotic symptoms will be assessed with established and validated self-report and clinician-rated instruments. In children and adolescents, the primary measure of depression symptom severity will be the broadly used and extensively validated 16-item clinician-rated Children's Depression Rating Scale-Revised (CDRS-R). In addition, the self- and parent-rated versions of the widely used and validated 41-item Screen for Child Anxiety Disorders (SCARED) will be used to measure anxiety symptoms in children and adolescents.
In adults, the current severity of depression symptoms will be measured primarily with the 10-item clinician-rated Montgomery and Åsberg Depression Rating Scale (MADRS), an internally consistent scale that is highly sensitive to therapeutic change, and the self-report Quick Inventory of Depressive Symptomatology - Self-Rated (QIDS-SR), a widely used and extensively validated measure of depressive symptoms that closely matches DSM diagnostic criteria for depression. Optional and alternative measures of depressive symptom severity in adults will include the clinician-rated Hamilton Rating Scale for Depression (HAM-D) and Quick Inventory of Depressive Symptomatology - Clinician Rated (QIDS-C) and the self-report Beck Depression Inventory (BDI),60,54 , the PROMIS depression scale (http://www.nihpromis.org/) and the Clinically Useful Depression Outcome Scale (CUDOS). Anxiety in adults will be optionally measured with the self-report Beck Anxiety Inventory (BAI), Penn State Worry Questionnaire (PSWQ), Social Phobia Inventory (SPIN), Impact of Event Scale (IES), Clinically Useful Anxiety Outcome Scale (CUXOS) and Cognitive Behavioural Avoidance Scale (CBAS), the PROMIS anxiety scale (http://www.nihpromis.org/) and the Hospital Anxiety and Depression Scale (HADS).
Manic symptoms will be optionally measured with the clinician-rated Young Mania Rating Scale (YMRS) and the self-report Altman Self-rated Mania Scale (ASMS). Psychotic symptoms will be optionally measured with the clinician-rated Positive and Negative Symptom Scale (PANSS).
Severity of suicidal ideation and behaviour in children and adults will be measured with the paediatric and adult versions of the Columbia Suicide Severity Rating Scale (CSSRS).
Status assessment: functioning The investigators will assess current level of functioning in children and youth with the widely used and validated Columbia Impairment Scale (CIS), parent and youth report. CIS is a 13-item questionnaire that provides a global measure of impairment with scored ranging from 0 to 52. Higher scores indicate a greater level of impairment. A score of 15 or higher is considered clinically significantly impaired. In adults, the investigators will assess current level of general, role- and social functioning in adults with the widely used and validated clinician-rated 2-item Role and Social Functioning Scales (RSFS), the self-report 5-item Work and the Social Adjustment Scales (WSAS) and the self-report 11-item Lam Employment Absence and Productivity Scale (LEAPS).
New Measures New instruments have been developed to measure two domains of interest for which no adequate established measures were identified: positive mood-related functioning and acceptability of and preferences for pharmacological and psychological treatments.
Engagement with Life Scale - self report (ELS-SR) is a self-report questionnaire developed based on qualitative material including interviews with individuals at different stages of recovery from depression and clinicians treating depression. It captures domains of functioning identified as related to mood and relevant to the quality of remission from depression. These domains include: enthusiasm, happiness, socializing, creativity, motivation, energy, interest, pleasure, decision-making, sense of purpose, self-confidence, resilience and hope. For example, individuals clinically recovered from depression often noted that before getting depressed they had been keen to try new things and associated this with being well, therefore, the measure includes items related to wanting to try new things to do and starting new activities. At present, this measure is at a developmental stage of a 50-item bank. The investigators propose to collect responses from 400 participants to obtain psychometric data that will allow shortening the measure to optimize efficiency and avoid redundancy.
Engagement with Life Scale - clinician rated (ELS-C). This is a 16-item clinician-rated scale covering the same domains as the self-report version (enthusiasm, happiness, socializing, creativity, motivation, energy, interest, pleasure, decision-making, sense of purpose, self-confidence, resilience and hope). It is administered as a semi-structured interview with obligatory and optional questions and then rated according to provided anchor points. The added value of the clinician-rated version is a degree of objectivity added by the clinician's judgement. Like the self-report version, this scale will undergo psychometric analysis and is likely to be refined and shortened based on the results.
Treatment acceptability and preferences questionnaire (TAP-Q). Antidepressant medication and structured psychological therapy (such as cognitive behavioural therapy, CBT) are comparably effective for treating depression and related conditions. Many individuals have strong preferences for either pharmacological or psychological treatment, but there is no established way for measuring such preferences. Pre-existing attitudes are likely to influence outcomes of treatment trials comparing psychological and pharmacological treatments. Therefore, the investigators have designed a measure that starts by providing brief descriptions of typical first-line pharmacological and psychological treatment, then asks the individual about the acceptability of each treatment, about any barriers to acceptability, whether there is a preference for one treatment modality over the other and whether the participant would be willing to receive one or the other treatment by chance (randomly). The investigators have obtained pilot data from 60 participants, who have found the measure clear and easy to complete. The investigators propose using this measure with 400 individuals with depression to carry out psychometric analysis and establish the rates of treatment preferences in Nova Scotia.
Patient-centered outcomes While the clinician may be most concerned with the reduction of symptoms or achieving remission, each individual with depression may have her/his own goals and expectations from treatment: e.g. for one person a good outcome may be returning to a busy career, while for others it may be the ability to read a novel or enjoy time with grand children. A patient-centred outcome (PCO) recommended by individuals with lived experience of depression aims to measure achievement of such personalized treatment goals. PCO will be personalized in content (individual's most valued treatment goal) and standardized in scale (0-100% goal achievement). PCO will be developed collaboratively with each patient in a goal-setting session where the patient will name her/his most valued outcome and with the help of an interviewer will create specific anchors for what conditions should be fulfilled for the personally valued outcome to be 25%, 50%, 75% and 100% achieved.
Physical health and side-effects of medication Unintended adverse reactions to medication are often a concern expressed by people living with mental illness. Therefore, the investigators will offer measurement of adverse reactions to medication ('side-effects') with several self-report scales including the Toronto Side Effects Scale (TSES), the Antidepressant Side-Effect Checklist (ASEC) and the Sex Effects Scale (SexFX). The three scales complement each other: TSES has the broadest range of symptoms, ASEC adds additional antidepressant-specific side effects and assesses the probability of association with an antidepressant, and SexFX assesses sexual side effects that are common but not well covered by the other scales. In addition, the database will allow recording routine health indicators including body weight, height, waist circumference, blood pressure, C-reactive protein (a measure of systemic inflammation and cardiovascular risk), triglycerides, cholesterol, glucose, insulin, and glycated haemoglobin (measures of glucose metabolism). These indicators will allow evaluating effects of medication on body weight, metabolic status and cardiovascular risk.
Data linkage The registry will have the capacity to be linked with public health and population databases using health card numbers. For participants who provide consent for data linkage, this will allow accessing additional information on health service utilization, prescriptions and diagnoses with no additional assessment burden. In addition, this will allow obtaining prospective information on longevity and causes of death and information about participants who are lost to follow-up.
Release of research data for clinical use To fulfill the mission of CDRIN to use research data for improvement of clinical care and following on suggestions from people with lived experience of depression, the registry participants will be able to choose that health-related information kept on them in the Registry be sent to their clinicians. This may provide additional information, save time, spare participants answering the same questions repeatedly, streamline clinical care and enhance evidence-based practice. Information will only be passed to clinicians on explicit (verbal or written) request of the participant. The information passed on to clinicians will take the form of a single sheet concisely summarizing the diagnostic and outcome information on the participant obtained so far as well as tips about possible treatments and community resources. Mental-health communication specialists (e.g. Healing Tree Communications) will review the form passed on to clinicians to maximise its usefulness and minimize the risk of passing on unhelpful or confusing information. External communication specialists will only provide guidance on the design and format of the generic feedback template. These communications specialists will not access any information on individual participants (identified or de-identified).
Potential benefits Unlike most ongoing research projects, the CDRIN Maritimes Registry will become a resource for clinical care and the participants are likely to personally benefit from their participation in the project. Participants will be able to request that the information collected about them in the Registry be sent to clinicians they are going to see. Participants will also be sent information about existing community resources, such as peer support and vocational services.
The study involving reliability of assessment instruments will not directly benefit the participants, but will ensure that instruments used to assess the diagnoses and severity of depression and related mental disorders are reliable and that different psychiatrists and other clinicians agree when making the diagnosis or measuring the severity of a disorder. More accurate measurement will improve the fit between individual's problems and prescribed treatment and will facilitate evidence-based practice. Training of health professionals in the use of reliable diagnostic and assessment instruments is known to improve the quality of clinical care. Routine systematic collection of outcome information will also facilitate the evaluation of specific treatment procedures, service changes and improvement initiatives. This will improve the quality of care for depression and other mental disorders in the medium to long term.
Potential harms All proposed procedures are safe and carry very low-risk. The potential harms include the burden of repeated interviews, upsetting content of diagnostic interviews and inadvertent breaches of confidentiality.
Burden of repeated interviews. The diagnostic interviews can be time-consuming, lasting typically 90-120 minutes. Repeating the same interview twice within a short time can be perceived as boring and it may put a strain on the participants' time resources. The investigators take the participants' time limitations seriously and the investigators take the following measures to limit the negative impact of interview burden. The investigators inform each participant in advance about the likely duration of interviews and emphasise that they are free to decide to participate or not in the interviews and that opting out of the repeated interview does not change their participation in the Registry. The investigators minimize the time required for interviews by efficient preparation and interviewer training. The investigators reimburse participants for their travel expenses and time spent with the repeated interviews.
Upsetting interviews. The diagnostic interviews and assessment instruments inevitably ask questions that can be upsetting for some participants. These include questions about the symptoms experienced in the past, suicide attempts, hospital admissions and traumatic experiences. Evidence shows that asking such sensitive questions is not harmful and that participants prefer to discuss these personally important issues with sensitive professionals rather than avoiding them. The investigators will minimize the upsets and provide a containing environment with the following measures. All staff conducting the interviews will have clinical experience and be able to contain upsets in a sensitive and responsible way. At the beginning of each interview, it will be emphasised that participants can refuse to answer specific questions or can chose to discontinue the interview if they find it too upsetting. An on-call system involving the investigators will be in place to bring in a psychiatrist if a participant is judged to be too upset or if risk to self or others is judged to be acutely increased. Most investigators are licensed psychiatrists or clinical psychologists, can attend at short notice, provide crisis intervention and arrange a referral if appropriate.
Breaches in confidentiality. In the course of this study participants will provide information that is highly sensitive since it concerns private experiences, information about mental health and personal information including contact details. The investigators will maintain the highest standards of confidentiality and minimize the potential harms from any breaches of confidentiality in the following ways. First, information that directly allows identification of subjects [names, addresses and similar] will be encoded using alphanumeric codes [the investigators will not use initials]. Collected information will be kept in a secured relational database in anonymized form with only the alphanumeric code identifying each participant. Personal information, contact details and their relationship to the alphanumeric code will be kept separate from the remainder of the data and secured with an even higher degree of security [kept in a password-protected file on a computer only accessible to the principal investigators and the study coordinators]. Even the anonymized data will only be accessible to study investigators and authorized staff of Dalhousie University, Capital District Health Authority and the Izaac Walton Killam (IWK) Health Centre, who are directly linked with the study and are familiar with principles of confidentiality and personal data protection. Information necessary for collaborative studies will only be made available to known researchers from academic institutions on signing a transfer agreement that includes binding statements on data confidentiality. All investigators will be required to keep study related data in encrypted/password-protected files on password-protected computers only, and take all precautions to prevent access by unauthorized individuals. No personally identifiable information will be made publicly available without separate REB approval and the participant's consent.
Balance of benefits and harms With the above outlined measures to minimize foreseeable potential harms, the investigators believe that the minimal risk of harm to participants will be outweighed by the expected direct benefits to participants, benefits to society and the advancement of knowledge stemming from the proposed research.
Future directions The investigators envisage that the Maritimes Depression Registry will become a long-term resource that will facilitate clinical research efforts and pioneer the iterative use of research information for improvement of clinical care and services. The assessment instruments with demonstrated reliability will become part of assessment for both research and clinical purposes. Specific instruments for assessing the positive aspects of mood and functioning and treatment preferences will fill in gaps in our assessment instruments. The registry will become a platform for large-scale low-cost clinical trials. The registry infrastructure will serve as a platform to develop and assess components to be incorporated in EHR systems that are expected to be implemented between 2019 and 2021.
Inclusion Criteria:
Exclusion Criteria:
jessica.toombs@nshealth.ca1-902-473-5313