Background:
Adolescents are vulnerable to common mental disorders
(CMDs) which are a leading cause of death and disability for this group, in
India. There are around 250 million adolescents in India.
Depression and self-harm account for a major share of the burden of death
and disability in this age group. A study from rural India
found that suicide rates amongst adolescents are amongst the highest in the
world (148/100,000 and 58/100,000 for females and males, respectively). The
intervention being tested involves an anti-stigma campaign and an electronic
decision support system (EDSS) platform that allows for identification,
diagnosis and treatment of CMD by doctors.
Aim and Hypotheses
The study aims to test clinical effectiveness and
implementation strategies to identify and reduce depression and suicide risk
among adolescents living in urban slums. We hypothesise that:
1. a community-based anti-stigma campaign
will lead to significant improvements in community behaviours toward
adolescents with common mental disorders; and
2. a mobile device-based
decision support system will improve the treatment of adolescents at high risk
of CMDs (which is defined as stress, depression and increased risk of
self-harm/suicide for this project) and lead to higher remission rates from
depression and reduced suicide risk.
Study Design
There are two phases
to the study.
(1) Optimisation
of anti-stigma and mhealth intervention: –
a. Anti-stigma
component: System level interventions that reduce stigma, and upskill primary
health care workers, including doctors to identify and treat people with mental
disorders by providing training and decision support tools, will be used to
improve mental health of adolescents.
b. mHealth
component: SMART Mental Health intervention 17 will be adapted
for adolescents (10-19 years) living in urban slums, in Vijayawada and Delhi,
to determine whether this strategy will increase remission rates for adolescents
with depression, other significant emotional or medically unexplained
complaints, and increased risk of self-harm/ suicide. Previous experience
of using mHealth and task sharing between physicians and non-physician health
workers (NPHWs), will be expanded to adolescents in slums. Community women volunteers would be chosen from the slums
and would have similar education level as Accredited Social Health Activists
(ASHAs). They will be trained on basic knowledge about mental health, stigma
and care of individuals with stress, depression, increased suicide risk.
(2) Effectiveness
testing – A cluster randomised trial with slum clusters as the
unit of randomisation. The trial will test clinical effectiveness and
implementation strategies across two geographical locations - Delhi and Andhra
Pradesh in India. About 69600 adolescents from both sites will be screened to
develop two cohorts of adolescents: those who are at high risk of CMDs and
those at low risk. Both cohorts will provide data on the stigma reduction
programme and facilitate understanding of the impact of the campaign on those
at high risk of CMD versus those not at high risk. Data from the high-risk
cohort will provide information on the effect of the mHealth programme over a
period of 12 months. A detailed process and economic evaluation of the trial
will be conducted. For participants in the intervention arm, the mHealth system
will capture health services usage using analytic data to assess intervention
fidelity. This will be followed by data analyses and dissemination
through public engagement meetings and policy symposia.
Study Duration: The total study duration is three years:
Phase 1 (intervention optimization and regulatory
approvals) – 1-12 months
Phase 2 (cRCT and post-trial assessment) - 13-36 months
Baseline Data collection:
All individuals included in both the high-risk and a
randomised cohort of low risk adolescents will have a detailed assessment by
interviewers where they will be asked questions about their socio-demographic
and health characteristics, mental health awareness and experience of
stigma/discrimination by family,community, friends,peers etc, social support,
stressors, treatment history, past history, alcohol and substance use, and
interpersonal violence
Randomisation:
Following baseline data collection, randomisation will be conducted at the
level of the ward/block. Allocation of ward/block level in a 1:1 ratio to
intervention or control will use a central allocation sequence and will be
stratified by geographic region and number of wards/blocks. The wards/blocks
selected under each intervention and control arms will be non-contiguous to
avoid contamination. Random allocation forward/blocks will be performed
using SAS enterprise guide 7.1 or later version by PROC plan procedure.
Interventions:
The core components of the intervention will be:
(1) an anti-stigma
campaign to improve community behaviours toward adolescents with CMDs; and
(2) implementation of
the mobile device-based decision support system to improve the treatment of
adolescents at high risk of CMDs and intended to lead to higher remission rates
from depression and reduced suicide risk.
Outcomes:
Anti-stigma component (combined
high and low risk cohorts):
Primary:
1. Change
in mean behaviour scores at the end of the trial using the KAB scale
Secondary
1. Change in mean knowledge and attitude scores and change
in stigma perceptions of adolescents as assessed by Barriers to Access to Care
Evaluation- Treatment Stigma (BACE – TS) Subscale
2. Change from baseline in mean stigma scores across both
high and low risk cohorts
2. mHealth component (high risk
cohort)
Primary
1. Proportion of high-risk achieving remission
(defined as all of the following: PHQ-9 <5, and suicide risk score
<2) at end of trial (12 months after randomisation).
Secondary
1. Standardised mean
difference in PHQ-9 scores at end of trial
2. Proportion at high risk of CMDs
(stress, depression and increased risk of suicide) who have visited a doctor at
least once between randomization and the end of the trial.
Sample size considerations
In
the combined high-risk and non-high-risk population: Assuming a
conservative ICC of 0.1 (0.01 in our pilot and 0.04 in similar studies) 6,15,
20 % loss to follow-up and a 2-sided significance level of 0.05, 60 clusters
and a mean cluster size of 66 per cluster (33 high-risk and 33 non-high-risk)
will provide more than 90% power to detect a standardised mean difference of
0.3 in mean behaviour scores between the intervention and control arms at 12
months. Assuming baseline mean behaviour scores of 2 (SD 1) and a 20% improvement
in the control arm based on pilot and published data,6 this corresponds to
12-month behaviour scores of 1.6 (20% improvement) and 1.3 (35% improvement) in
the control and intervention arms respectively. This sample size also provides
80% power to detect differences of effect size of 0.5 for subgroup analyses
considering half of the numbers belong to the high risk and low risk groups.
The individual samples for high-risk and low risk allows us to measure changes
in the behavioural score separately across both these groups.
For people with or at high risk
of CMDs: Assuming a 50% remission
rate in the control arm at 12 months based on published data, and
30 clusters per intervention arm, at least 27 subjects per cluster are needed
to detect a 15% absolute improvement in the intervention arm (65% intervention
vs 50% control) at 12 months with 90% power. This assumes a 2-sided
significance level of 0.05 and an intra-class correlation coefficient (ICC) of
0.1 based on pilot data and recent (unpublished) analyses. For
secondary outcomes, this sample size will also provide 86% power to detect a
standardised mean difference of 0.3 in PHQ-9 scores considering an ICC of 0.1.
The hypothesised effect size is consistent to that published in a recent Indian
study on provision of mental health services using lay health workers in
community settings.
To account for up to 20% of participants lost to
follow-up, we will aim to enrol at least 33 participants per cluster for a
total sample size of 1980 adolescents (60 clusters × 33 adolescents). We
anticipate a prevalence of approximately 3% and will therefore aim to screen
about 1,100 adolescents in each cluster. Further assuming that 5% of eligible
adolescents will refuse to participate, we aim to screen 1,160 adolescents in
each cluster i.e. a total of 69,600 adolescents (60 clusters × 1,160
adolescents). The sample has more than 80% power at 2α = 0.05 to assess sex disaggregated
estimates for both the primary outcomes, assuming equal proportion of male and
female in the high-risk and low risk cohorts. Research suggests that among
adolescent’s depression is equally distributed in males and females.
Data collection
Assuming a prevalence of 3% of adolescents with CMDs
(based on published and Andhra Pradesh data), we anticipate that around 34800
adolescents per region will need to be screened to achieve the required sample
size. Both ‘high-risk’ and ‘low-risk’ cohorts will be re-interviewed at 3, 6,
12 months. Outcome data will be collected by trained interviewers, blinded to
intervention allocation. Blinding will be done by recruiting staff who will be
trained to collect only this data and who has not been involved in any of the
previous phase of the study. Participants will be interviewed for approximately
one hour in their house and the same questionnaires will be administered at
follow-up, including questions on mental health services use and quality of
life.
Quantitative data on stigma perceptions related to mental
health and depression/anxiety will be collected prior to the randomisation
(Time 0) and then subsequently at 3, 6, 12 months of the intervention.
Time - 0 month (pre-randomisation): The names of those at high-risk will be shared with
trained interviewers who will then administer a detailed questionnaire that
will enquire about sociodemographic characteristics, treatment history, past
history of any mental illness, family history of any mental illness, social
support from friends and families, stressful events experienced in the last one
year, history of any comorbid major physical illnesses.
Time
– 3, 6, 12 months (intervention phase): Data will be collected from
all high-risk and a random sample of low risk adolescents across all the
wards/blocks by trained interviewers. Questionnaires will administer the PHQ9,
KAB and BACE-TS at these time points.
Process and Economic evaluation
Processes will be monitored continuously throughout the
intervention phase. A formal evaluation will be implemented using Michie’s
behaviour change theory as the over-arching framework to guide our
understanding of the intervention implementation and impact. This theory
assesses the capability, opportunity and motivation of adolescents, community
women volunteers/NPHWs and doctors to engage with the intervention. We will
take a case study approach in which a purposive sample of clusters will be
selected. Qualitative data about the experiences of the community participants,
UHC worker, doctors, and other key stakeholders will be gathered through focus
groups and in-depth interviews. For participants in the mHealth intervention
arm, the mHealth system will capture health services usage analytic data to
assess intervention fidelity. A mixed methods analysis will be conducted like
the approach taken in earlier trials and following MRC guidelines. A
trial-based and a modelled evaluation of long-term costs and outcomes will be
conducted from a health system perspective.
Data Analyses
Suitable descriptive and analytic statistics will be conducted,
and models will be developed to understand factors associated with the outcome
variables. Qualitative data will be analysed using appropriate techniques to
identify key concepts