| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH |
|
|
|
|||||||
|
|||||||||
TRACK to:
|
|
Electronic letters published:
![[Read Comment]](/icons/misc/sectionDOWN.gif){kind=icon}
Response to question on alternate statistical method
Response to question re change in OR with adjustment
questions about the data analysis
Commentary on Beliefs and Attitudes Associated With the Intention to Not Accept the Diagnosis of Depression Among Young Adults
Using the Internet to understand non-acceptance of a depression diagnosis
|
|
|||
|
Bejjamin W Van Voorhees, Chicago Physician - author
Send response to journal:
|
The demographic variables were minor prectors on their own right. I subsequently conducted a stratified analysis of attitudes by quartiles of CESD score. Although I did not do this with every attitude variable, there did not appear to be an important relationship between symptom level (CESD) and attititudes or social norms. The main effect of gender may have been through past behavior as described in another comment. We evaluated this with elimination studies described in the other comment response. Although structural equation modeling would have provided an elegant way to explore these relationships further, the apparent modest relationships between the internal and external variables did not seem to justify this. Additionally, we would have had to report coefficients that would be difficult to interpret for clinicians. Competing interests: None declared |
|||
|
|
|||
|
Bejjamin W Van Voorhees, Chicago Physician - author
Send response to journal:
|
With regard to change in OR for male gender, para 1 on page 43 states that the addition of past behavior and family history explained most of the change. We conducted analyses to examine which variables explained the change in OR for ethnicity (which did not end up in the final version due to space constraints). With regard to ethnicity and gender, we found that past behavior was the most important variable. In short, med and ethnic minorities tended to have less exposure to past treatment and perceived those experiences to be less helpful than white woman in this sample. However, if you sequentially add each variable to the model, attitudes and social norms also have an incremental impact. There are some important variations by ethnicity, African Americans are more likely to endorse the statement that "prayer can heal depression". Competing interests: None declared |
|||
|
|
|||
|
Jason Chao, Cleveland, OH USA Family Practice, Case Western Reserve Univ.
Send response to journal:
|
We recently discussed this article in a journal club with some first year medical students. Some questions arose about the data. In Table 2, the subgroups for each factor add up to close to 100% except for Treatment preferences, which only add up to 86.8%. In Table 3, the odds ratio for several of the factors (gender, specific ethnicities) change not only in magnitude after adjustment, but also in direction – from greater than one to less than one. So before adjustment, the factor is correlated with greater Intention, but after adjustment, the same factor is correlated with less Intention. There is no comment on this extreme change after adjustment in the text. According to Figure 1 explaining the theory of reasoned action, external factors act on Intention, mediated through Internal Factors, Subjective Norms and Past Behavior. Would a different statistical analysis such as path analysis be more appropriate? Competing interests: None declared |
|||
|
|
|||
|
Elizabeth H.B. Lin MD, MPH, Seattle, USA Primary Care Physician/Scientific Investigator, Group Health Cooperative
Send response to journal:
|
In this issue of Annals of Family Medicine, Dr. Van Voorhees and colleagues’ study adds much to our understanding of factors associated with intention not to accept depression diagnosis among an under-studied population. Applying theory of reasoned action, the authors surveyed a large number of 16-29 year olds using the Internet. Compared to primary care populations, this study population included more young men and those with no prior depression treatment. This investigation found that specific beliefs and attitudes account for much of the variance (86%) in the intention not to accept depression diagnosis. Findings related to lower intention to accept depression diagnosis such as - negative beliefs of biological basis of depression, ineffectiveness of medications, and preference of counseling, have practical implications for interventions. Tailoring interactive education materials according to individual beliefs about cause and treatment expectations can help depressed persons gain better understanding of depression and engage in evidence-based treatment. It is interesting that preference for counseling was negatively association with intention to accept depression diagnosis. This finding can guide us to create educational information that demonstrate benefits of psychotherapy and how these benefits link closely to biologic aspects of depression. In this elegant survey, Dr. Van Voorhees and colleagues made important contributions in our knowledge toward a more patient- centered approach for depression, and provide guides for clinical applications. Respectfully submitted by Elizabeth HB Lin MD, MPH Competing interests: None declared |
|||
|
|
|||
|
Joshua Fogel, Brooklyn, NY, USA Assistant Professor, Department of Economics, Brooklyn College
Send response to journal:
|
In the article by Van Voorhees et al. (1), the challenging issue of understanding why young adults refuse to accept a diagnosis of depression is explored. One interesting approach was the method of obtaining the data. Unlike many surveys or interviews where anonymity is assured, there is often the subtle belief from participants that possibly what they respond to or write on the survey is being judged by the interviewer or the one collecting the completed survey. Especially regarding sensitive topics, there is sometimes a hesitation by individuals to completely and truthfully respond to the question. In this Internet survey approach, the participants completed anonymous surveys posted on a website without these surveys being linked to the text of an e-mail address. In my clinical experience, I have seen that certain sensitive questions that are asked by using a computer, rather than a face-to-face interview, elicit more honest answers than what the patient would have told me face-to-face. This concept is also validated by the scientific literature where Joinson has extensively studied this topic (2-3). He concludes that more self-disclosure occurs when individuals participate in research over the Internet as compared to more traditional approaches of face-to-face interviews or paper surveys. Besides this Internet approach where more self-disclosure may have occurred, Van Voorhees et al. (1), incorporated the theory of reasoned action, as the basis for testing a variety of possible reasons for understanding why young adults refuse to accept a diagnosis of depression. The sample size of 10,962 is sufficiently large enough to allow one to be comfortable that there was adequate statistical power to concur with the results. This study by Van Voorhees et al. (1) will hopefully help clinicians understand the relevant issues regarding whether young adult patients will accept a clinician’s diagnosis of depression. 1. Van Voorhees BW, Fogel J, Houston TK, Cooper LA, Wang, N-Y. Beliefs and attitudes associated with the intention to not accept the diagnosis of depression among young adults. Ann Fam Med. 2005;3:38-45. 2. Joinson A. Social desirability, anonymity, and Internet-based questionnaires. Behav Res Methods Instrum Comput. 1999;31:433-438. 3. Joinson A. Knowing me, knowing you: Reciprocal self-disclosure in Internet-based surveys. Cyberpsychol Behav. 2001;4:587-591. Competing interests: None declared |
|||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH |
TRACK: