Initiative Investment Analysis for DepressionCare

Case Description

You have been assigned a new position in the agency DepressionCare, a non-governmental organization with the goal of expanding and improving diagnosis and treatment of depression in a disadvantaged community setting. You have limited resources, but numerous options. You can invest all your resources in (1) a public campaign to de-stigmatize treatment seeking for depression, (2) an initiative to improve detection and treatment of major depression in primary care settings, or (3) an initiative to improve the quality of care for patients with depression treated in specialty mental health clinics and hospitals.

Objectives

Depressive disorder currently stands as the leading cause of disability-adjusted lost-years (DALYs) in high- and middle-income countries. (Ustiin et al. 1999) Levers to combat the incidence and prevalence of depressive disorder span from promotion and prevention to treatment and maintenance. (Mojtabai 2013) DepressionCare’s investment options span across these levers. Given this and DepressionCare’s limited resources, it is important to choose to invest in an initiative that will have the highest impact.

As stated in the case description, the goals of DepressionCare are “expanding and improving diagnosis and treatment of depression”. Further, the initiative that we invest in will be working “in a disadvantaged community setting.” We can assume that this likely means working with a population exhibiting higher poverty rates, lower education levels, and a more diverse set of ethnic backgrounds if we assume this is taking place in the United States.

Decision Criteria

To evaluate the alternatives, there are a number of criteria that we can use. For the purposes of this consultation, we will base our decision on the following four factors.

Potential scale of treatment expansion

As outlined in the case description, DepressionCare is interested in “expanding…diagnosis and treatment of depression”. Some factors affecting “reach” include:

• The total number of individuals with depression reached per dollar spent
• How easily our efforts can be scaled (either through empowering others to replicate our initiatives or otherwise)

Ability to improve treatment

Additionally, DepressionCare is interested in “improving diagnosis and treatment”. The following metrics will be used to assess this factor:

• Number needed to treat (NNT) of intervention type
• NNT discounted by cost to deliver intervention (i.e. NNT x Cost / Intervention)

Likelihood of successful project execution

Ultimately, it is important to be able to execute on the project. If we are not able to pull through and execute, then even the greatest of ideas would be sought in vain. This will be evaluated using the following metrics:

• Overall costs - This will be done assuming an average North American NGO budget size of $1M, which is much higher than the average NGO budget for non-North American NGOs
• DepressionCare employee competencies - Assuming the employee base comprises those with a knowledge base of someone who has passed this course with distinction

Extent of social and economic impact

While our initiative may be able to affect a certain number of individuals, the ability to add more to the economy or raise additional funding for more work is another factor that we will want to consider. This will be done by evaluating the following:

• Increased productivity and employment opportunities from those affected (tax revenues gained from this increase)
• Number of additional dollars potentially raised for future initiatives helping the cause

Evaluation of Initiatives

Given the variety of ways in which the three presented initiatives could be executed, the choice of initiative is necessarily biased to the perspectives, competencies, and creativity of DepressionCare employees. Nonetheless, that does not preclude us from briefly assessing the initiatives based on our chosen criteria. Each factor will be evaluated on a scale from one to five on impact level, with five being the most impactful, and one being the least impactful. Scores will be summarized and totaled at the end, with each initiative being scored on a scale from 1-20. That score will be the primary point we will use to determine which initiative to invest in.

Initiative I: Public campaign to de-stigmatize treatment seeking for depression (18/20)

In this section we will evaluate the merits of Initiative I based on the aforementioned factors.

Potential scale of treatment expansion (5/5)

• The stigma associated with mental illness prevents two- thirds of those affected from seeking help. (Mental Health Commission of Canada, Mental Health, and Canada 2012)
• This means that the majority of non-treated individuals are currently facing stigma as a barrier
• There is a great deal of opportunity to empower others to play a role in the campaign and therefore add leverage to efforts

Ability to improve treatment (3/5)

• Universal prevention techniques such as those used by public campaigns are, based on meta-analyses of studies, “less effective than selective and indicated interventions.” (Cuijpers et al. 2008) However, in the study where these findings were reported, “universal interventions were examined in only two studies, and pooling of these studies or a comparison with selective or indicated interventions (???) not very meaningful,” (Cuijpers et al. 2008)
• If selective or indicated campaign interventions are used, NNT could be brought down to 16 or 17. (Cuijpers et al. 2008)
• Further, as NNT for interpersonal therapy is 5, including this as a part of the education process could make the interventions even more impactful (Cuijpers et al. 2008)

Likelihood of successful execution (5/5)

• The primary costs of executing on a public campaign have decreased with the increased adoption of social media and online technologies. Therefore, it is likely that DepressionCare could have a measurable impact within the allocated budget.
• Given the competencies of the one DepressionCare employee surveyed for this analysis, it is clear that, while the organization is full of strong and high-potential Coursera students, it lacks qualified MD’s and/or R&D professionals. However, for the purposes of a public campaign, the organization is well equipped and full the the requisite zest and zeal to pull it off!

Extent of social and economic impact (5/5)

• A public campaign brings with it the ability to solicit funds for other initiatives. (e.g. Movember with its Men’s Mental Health initiative)
• Decreased stigma could lead to greater employment opportunities for those facing depression.

Initiative II: Initiative to improve detection and treatment of major depression in primary care settings (12/18)

In this section we will evaluate the merits of Initiative II based on the aforementioned factors.

Potential scale of treatment expansion (2/5)

• There is truly a great argument for the improvement of detection and treatment in primary care settings. While 78% of those seeking treatment for major depression do so through primary care (vs. 18% through mental health specialists), only 25% of patients end up receiving any form of treatment. (Mojtabai 2013)
• Additionally, clinical diagnosis accuracy rates are incredibly low, with specificity estimates of 50% based on recent studies. (Mojtabai 2013)
• However, the populations addressed here are among those who have already sought treatment. Given that the majority of those suffering from depression still do not actively seek treatment, the scale of potential impact is less than a successful campaign given optimal circumstances.

Ability to improve treatment (4/5)

• Let us assume that, in order to develop novel methods for detection and treatment of major depression, we must have the following: - Strong research staff - Large amount of resources - Knowledge base that materially improves upon existing well-funded research
• With that assumption in place, it would be relatively difficult and expensive for use to truly move the needle on the efficacy of depression treatment through this initiative
• As such, the NNT of this initiative makes it relatively unfeasible given our previous assumptions

Likelihood of successful execution (2/5)

• Given our DepressionCare staff competencies, the likelihood of developing a novel innovation to improve detection and treatment of major depression to perform better than a coin flip would be both expensive and difficult

Extent of social and economic impact (2/5)

• Aside from raising research dollars to investigate methods for innovation, there is likely no near-term material social or economic benefit from pursuing this option
• Increased employment and productivity gains from this are relatively small compared to the public campaign

Initiative III: Initiative to improve the quality of care for patients with depression treated in specialty mental health clinics and hospitals (14/20)

In this section we will evaluate the merits of Initiative III based on the aforementioned factors.

Potential scale of treatment expansion (4/5)

• Given that the number of individuals who are treated in specialty mental health clinics is much lower than those treated through primary care (Lecture Notes), that segment is lower than the potential audience of a public campaign.
• However, within hospitals, major depression is a serious co-morbidity within many patients. Patients suffering from neurological diseases (e.g. Multiple Sclerosis, Dementia, Epilepsy) have co-morbid depression at rates two to five times higher than the normal patient population. (C. Thielscher, Thielscher, and Kostev 2013)
• Scalability of this initiative would be rather limited and resource intensive, unless paired with a public campaign. This would be due to the number of trained clinical staff necessary to help implement and disseminate best-practices.

Ability to improve treatment (4/5)

• Ability to improve treatment is relatively high, particularly for treating co-morbid depression within hospital patients. A collaborative care model implemented in Dutch clinics showed low NNT figures of 2-3. (Huijbregts et al.)
• The ability to improve treatment within mental health clinics, however, may be more difficult, and would require more research and development.

Likelihood of successful execution (3/5)

• Again, given DepressionCare staff competencies, successful execution could prove difficult
• The likelihood of success may be increased if a model like that in the Dutch study could be used within a an outreach campaign to health providers

Extent of social and economic impact (3/5)

• Research dollars could be raised potentially, but are likely more limited than for the campaign given the scope
• Increased adherence to medication could help with the recovery process and therefore allow for more productivity from those who are able to heal further
• However, many of those in hospitals suffering with co-morbid depression may be older, and therefore their lifetime earnings may have a lower impact than targeting youth

Investment decision

Given the points listed above, Initiative I has the highest score based on our criteria, with 18 points out of a total of 20. The next section of the report will focus on some of the details of the program, and particularly how we can design the program to have maximal impact.

Description of Chosen Initiative

Anti-Stigma Campaign for Employers of Youth and At-Risk Individuals

Former director of the World Health Organization’s Division of Mental Health, Norman Sartorius, stated at a recent conference that “stigma has actually increased over the past decade”, and that “people other than psychiatrists should lead anti-stigma programs - as determined by their strength of personality, not their academic credentials.” (Mental Health Commission of Canada, Mental Health, and Canada 2012)

However, how targeted should we be with our campaign? Sartorius additional argues that the mental health community should design campaigns with “targeted, specific messaging and for anti-stigma programming to be part of routine, ongoing work.” (Mental Health Commission of Canada, Mental Health, and Canada 2012)

Given that “ethnic minorities, youth, men, and those in military and health professions were disproportionately deterred by stigma”" (Clement et al. 2014), we will focus our efforts on male youths in primarily ethnic communities.

Further, since employers and managers with a higher a job grade are more likely to hold stigma than employees with a lower job grade (Wang 2011), and since employer stigma is a major barrier for employment and productivity in the community, we will also focus on employers of male youths in ethnic communities.

Group interventions were found in a recent study to have a lower NNT compared to individualized interventions. (Cuijpers et al. 2008). Therefore our campaign will involve a strong component of meeting with groups of employers, and facilitating workshops. This could be accomplished by meeting with Chambers of Commerce, or by meeting with teams of managers at companies.

A study on drivers of mental health stigma within gender populations concluded that “mental health education and promotion should clarify misconceptions about causes, treatments and risk factors for depression.” (Wang et al. 2007) As such, the campaign will focus on depression education for our target audience.

Hopefully we can develop training and marketing material to help educate managers about the realities of depressive disorder, why it should not be a barrier to employment, and how they can best accommodate any of their depressed employees.

---

Outcomes Considered

The outcomes and factors considered were outlined in the Decision Criteria section in the previous answer; however, for the purposes of completeness, they are repeated again below:

Potential scale of treatment expansion

As outlined in the case description, DepressionCare is interested in “expanding…diagnosis and treatment of depression”. Some factors affecting “reach” include:

• The total number of individuals with depression reached per dollar spent
• How easily our efforts can be scaled (either through empowering others to replicate our initiatives or otherwise)

Ability to improve treatment

Additionally, DepressionCare is interested in “improving diagnosis and treatment”. The following metrics will be used to assess this factor:

• Number needed to treat (NNT) of intervention type
• NNT discounted by cost to deliver intervention (i.e. NNT x Cost / Intervention)

Likelihood of successful project execution

Ultimately, it is important to be able to execute on the project. If we are not able to pull through and execute, then even the greatest of ideas would be sought in vain. This will be evaluated using the following metrics: - Overall costs (This will be done assuming an average North American NGO budget size of $1M, which is much higher than the average NGO budget for non-North American NGOs) - DepressionCare employee competencies (Assuming the employee base comprises those with a knowledge base of someone who has passed this course with distinction)

Extent of social and economic impact

While our initiative may be able to affect a certain number of individuals, the ability to add more to the economy or raise additional funding for more work is another factor that we will want to consider. This will be done by evaluating the following:

• Increased productivity and employment opportunities from those affected (tax revenues gained from this increase)
• Number of additional dollars potentially raised for future initiatives helping the cause

Measuring and Monitoring Outcomes

There are a number of potential measures that could be used to measure and monitor stigma. The following are just a few that are commonly used in practice. These have been collated from a presentation given by Dr. JianLi Wang in 2008.

• Social Distance - 7 items (Link & Cullen, 1983)
• Discrimination Questionnaire - 26 items (Corrigan et al., 2003)
• Self-stigma (Corrigan, 2004)
• Depression stigma scale - 18 items (Griffiths et al., 2004)
• The Stigma Scale - 28 items (King, et al., 2007)

(Wang and Lai 2008)

Outcome Measure Choice

For the purposes of our initiative, we can evaluate the outcomes by leverage the Griffiths scale as recommended and implemented by Wang in a series of Stigma research performed in Calgary.

The implementation of the Griffiths scale usually involves two steps: First, having subjects read a case study of someone exhibiting symptoms of depression, and then asking them to answer a set of questions to evaluate their stigma against the person outlined. The scale employs a series of questions, nine of which are listed below, to measure stigma.

Personal Stigma Items from Griffiths Depression stigma scale

1. People with a problem like X’s are unpredictable
2. I would not vote for a politician if I knew they suffered a problem like X’s
3. I would not employ someone if I knew they had a problem like X’s
4. People with a problem like X’s are dangerous
5. People with a problem like X’s could snap out of it, if they wanted
6. If I had a problem like X’s I would not tell anyone
7. A problem like X’s is a sign of personal weakness
8. X’s problem is not a real medical illness
9. It is best to avoid people with a problem like X’s so that you don’t develop this problem

(Wang and Lai 2008)

These questions are rated by the subjects on “a five-point Likert scale:” strongly agree, agree, neither agree nor disagree, disagree and strongly disagree." (Wang and Lai 2008) Therefore, we can assume that lower scores from the study participants would point toward lower stigma overall.

Sartorius from the World Health Organization recommends that “programs should not be measured by whether they change attitudes but behaviour.” (Mental Health Commission of Canada, Mental Health, and Canada 2012) Therefore, it will also be important to measure whether behaviours of employers toward male youths with depression are changed. This can be evaluating by measuring the change in the number of male youth with depression whom employers have hired before and after the initiative.

Outcome Measures to Exclude

Simply using number of individuals reached through the campaign and workshops may not be the most effective method of evaluating the effort. This is because, while a number of people may have been reached, this does not measure the change in behaviour or even change in attitude that Sartorius emphasized was important with these campaigns. (Mental Health Commission of Canada, Mental Health, and Canada 2012)

Study Design

For the purposes of this initiative, we can perform a prospective cohort study evaluating the measures outlined above within a randomly sampled group of managers participating in the workshops and campaign activities. The study will take the following steps:

1. Identify a set of employers to conduct the study with
2. Divide the employer groups into a control group and the study group, with whom we will conduct workshops and deliver campaign material about de-stigmatizing male youths in the workplace
3. Deliver the survey outlined above to a randomly selected subset of both the control group and study group
4. Get a baseline estimate of number of individuals with depression and male youth with depression employed by these groups
5. Conduct workshops with the study groups
6. After six months, and again after one year, repeat steps #3 and #4
7. Calculate the difference in results over the time points and evaluate differences between the control group and the study group

If the differences are high enough between the two groups, this could be an indicator that the campaign has been effective, and is worth scaling!

References

Clement, S, O Schauman, T Graham, F Maggioni, S Evans-Lacko, N Bezborodovs, C Morgan, N Rüsch, JSL Brown, and G Thornicroft. 2014. “What Is the Impact of Mental Health-Related Stigma on Help-Seeking? A Systematic Review of Quantitative and Qualitative Studies.” Psychological Medicine. Cambridge Univ Press, 1–17.

Cuijpers, Pim, Annemieke van Straten, Filip Smit, Cathrine Mihalopoulos, and Aartjan Beekman. 2008. “Preventing the Onset of Depressive Disorders: A Meta-Analytic Review of Psychological Interventions.” American Journal of Psychiatry 165 (10). Am Psychiatric Assoc: 1272–80.

Huijbregts, Klaas M.L., Fransina J. de Jong, Harm W.J. van Marwijk, Aartjan T.F. Beekman, Herman J. Adèr, Leona Hakkaart-van Roijen, Jürgen Unützer, and Christina M. van der Feltz-Cornelis. “A Target-Driven Collaborative Care Model for Major Depressive Disorder Is Effective in Primary Care in the Netherlands. a Randomized Clinical Trial from the Depression Initiative.” doi:doi:10.1016/j.jad.2012.09.015.

Mental Health Commission of Canada, World Psychiatric Association Scientific Section on Stigma, Canadian Human Rights Commission, Mental Health, and Public Health Agency of Canada. 2012. “A Report on the 5th International Stigma Conference.” Together Against Stigma: Changing How We See Mental Illness.

Mojtabai, Ramin. 2013. “An Introduction to Treatment of Depression.” Johns Hopkins University. https://d396qusza40orc.cloudfront.net/pmhdepression/lecture_slides/MajorDepression6A_TreatmentPix.pdf.

Thielscher, C, S Thielscher, and K Kostev. 2013. “The Risk of Developing Depression When Suffering from Neurological Diseases.” GMS German Medical Science 11. German Medical Science.

Ustiin, T Bedirhan, Jiirgen Rehm, Somnath Chatterji, Shekhar Saxena, Robert Trotter, Robin Room, and Jerome Bickenbach. 1999. “Multiple-Informant Ranking of the Disabling Effects of Different Health Conditions in 14 Countries.” Lancet 354 (111): 15.

Wang, JianLi. 2011. “Mental Health Literacy and Stigma Associated with Depression in the Working Population.” In Work Accommodation and Retention in Mental Health, 341–51. Springer.

Wang, JianLi, and Daniel Lai. 2008. “The Relationship Between Mental Health Literacy, Personal Contacts and Personal Stigma Against Depression.” Journal of Affective Disorders 110 (1). Elsevier: 191–96.

Wang, JianLi, Gordon Fick, Carol Adair, and Daniel Lai. 2007. “Gender Specific Correlates of Stigma Toward Depression in a Canadian General Population Sample.” Journal of Affective Disorders 103 (1). Elsevier: 91–97.