Depression Self-Management Assistance Using Automated Telephonic Assessments and Social Support

Objective:

To understand patient participation in interactive voice response (IVR) depression monitoring and self-management support calls and estimate the workload from clinical alerts based on patients’ IVR reports.

Study Design:

Observational study from program implementation in 13 community- and university-based primary care practices.

Methods:

Patients with depression were identified using electronic records and enrolled by telephone. Patients were asked to complete IVR assessments weekly; those with significantly improved symptoms had the option of completing assessments monthly. Patients could enroll with an informal caregiver who received automated feedback based on patients’ IVR reports. Clinicians received alerts regarding significant changes in the scores on the Patient Health Questionnaire 9-item depression measure, antidepressant adherence problems, or suicidal ideation.

Results:

A total of 387 patients were followed for 12,042 weeks. More than half (59%) opted to participate with a caregiver. Patients completed 68% of 7912 attempted IVR assessments. Assessment completion was unrelated to patients’ depressive symptoms and was higher among those who participated with a caregiver, were married, had more comorbidities, or reported missing a prior appointment. Assessment completion was lower when patients received monthly versus weekly assessment attempts. Clinical alerts were generated during 4.9% of follow-up weeks; most represented medication adherence problems (2.8%). Alerts indicating suicidal ideation were rare (0.2% of patient-weeks).

Conclusions:

IVR support calls represent a viable strategy for increasing access to depression monitoring and self-management assistance in primary care. These programs generate a manageable number of alerts, most of which can be triaged with limited physician involvement.

Am J Manag Care. 2013;19(11):892-900We found that an interactive voice response (IVR) monitoring and self-management program for depressed patients was a viable strategy for increasing access to depression management services in primary care.

• Most patients opted to participate with an informal caregiver, who received automated feedback based on the patient’s IVR reports.

• Patients completed 68% of 7912 attempted IVR assessments. Completion rates were similar regardless of patients’ depressive symptoms and were higher among those participating with a caregiver.

• The program generated a manageable number of alerts, most of which could be triaged by allied health professionals with limited physician oversight.

Roughly 7% of Americans have major depression each year,¹ and depression is expected to be the second-most burdensome disease worldwide by 2030.² Individuals with depression are more likely to have serious medical conditions,³ have twice the mortality rate as nondepressed people,⁴ and incur greater healthcare costs.⁵ Depression is common among primary care patients,⁶ especially those with chronic medical illnesses.⁷ Comorbid depression impairs self-care and worsens medical outcomes.⁷

Effective depression management is impeded by inadequate provider training, limited visit time, and few resources to monitor and support self-care.⁸ Cognitive-behavioral therapy is effective, but limited by the small number of providers and program cost.⁹ Antidepressant medications can reduce symptoms and recurrence risk, but adherence is often poor.¹⁰ For these reasons, routine depression screening is of little benefit without formal systems for follow-up.¹¹ Evidence-based depression care typically includes telephone care management to enhance recovery. While such programs may be cost-effective,^12,13 they can be unattractive to decision makers because of the sheer number of undertreated patients, staffing requirements,¹⁴ and frequent contact many patients need.^15,16

Mobile health services including interactive voice response (IVR) calls address these barriers to effective depression care management. Patients with a variety of chronic conditions will respond to IVR calls¹⁷ and provide valid and reliable clinical information.^18-22 Given their efficacy in other chronic diseases, IVR-based interventions may improve mental health outcomes.^23,24

Another strategy for improving depression management is to enhance patients’ social support. In 1 trial, trained laypersons led to a 65% remission rate for depressive symptoms compared with usual care (39%).²⁵ Patients with greater social support have fewer depressive episodes^26-28 and lower symptom severity, and most attribute their depression to insufficient support.^29,30 However, spousal caregivers are at risk for burnout from competing demands,^31-34 and few informal caregivers have the tools needed to systematically monitor patients’ mood and support their self-care.

Here, we describe the implementation of a primary care—based intervention in which depressed patients completed weekly IVR mood and self-care telephone assessments, and received tailored messages related to antidepressant adherence,mood monitoring, and behavioral activation. The intervention also primary care, and with informal caregivers. We examined variation in program engagement as defined by completion of IVR calls and the frequency and types of clinical alerts sent to primary care.

METHODSPatient Eligibility and Recruitment

Patients were enrolled between March 2010 and January 2012 from 13 university-affiliated and community-based primary care practices. Eligible patients had to have 2 primary care visits in the previous 2 years and 1 in the previous 13 months, and either an active depression diagnosis or an antidepressant prescription plus billing diagnosis of depression. Patients with schizophrenia, psychosis, delusional disorder, bipolar disorder, or dementia were excluded. Potential participants were mailed an introductory letter followed by a telephone call. After providing written informed consent, patients were mailed additional program information, including materials describing effective communication with informal caregivers and clinicians. Caregivers provided oral consent to receive feedback and suggestions based on their patient-partner’s IVR assessments. The study was approved by the university human subjects committee.

Program Features

Interactive Voice Response Calling System

The intervention provided (1) patients with IVR monitoring of depressive symptoms and medication adherence, including tailored information about self-care and when to seek health services; (2) clinical teams with actionable feedback about IVR-reported problems; and (3) informal caregivers with feedback about patients’ status plus guidance on supporting self-management. Each week an assessment was scheduled (ie, call-week), the system made up to 3 call attempts on up to 3 patient-selected day/time combinations. Call contents were developed with input from psychiatrists, primary care providers, and experts in IVR design and health behavior change. Details about the call contents and flow can be obtained by contacting the authors.

After patients verified their identity, their depression symptoms were assessed using the Patient Health Questionnaire 9-item depression measure (PHQ-9).³⁵ Medication adherence was assessed using a standard item: “How often during the past week did you take your depression medication exactly as prescribed?” Additional questions asked about days in bed due to mental health symptoms and perceived general health. Calls used tree-structured algorithms to present recorded queries and information, and lasted between 5 and 20 minutes. Weekly assessments were scheduled for weeks 1 through 6. Thereafter, patients with mild depressive symptoms (3 consecutive PHQ-9 scores of <10) were automatically given the option to reduce the frequency to 1 call-week per month. Patients were automatically given the option to revert to weekly calling whenever their PHQ-9 scores were 10 or higher.

Enhanced Social Support

Patients could participate with a family member or friend (ie, an informal caregiver) and completed the Norbeck Social Support Questionnaire³⁶ to identify the best candidate. Eligible caregivers needed to be at least 18 years old, to report no history of psychosis or cognitive impairment, and to agree to participate. Caregivers automatically received structured e-mails based on information reported during the patient’s IVR calls, with feedback about the patient’s status and tailored self-management support advice.

Clinician Alerts

As part of a series of meetings to gain input and review the IVR call contents, clinicians (physicians and nurses) in primary care and psychiatry defined thresholds for alerting primary care teams about urgent patient reports. Alerts were designed to be actionable, have low false-positive rates, and efficiently use human resources for follow-up. The following 3 conditions triggered alerts: suicidal ideation, poor medication adherence, and increase in depressive symptom severity.

Suicidal Ideation. Alerts were generated if patients reported they “had made a specific plan to harm [themselves] or end [their] life,” or that it was “somewhat likely” or “very likely” they would “harm [themselves] or end [their] life sometime over the next few days.” Patients reporting suicidal thoughts were instructed to talk with their doctor or mental health professional as soon as possible and were told that their clinician would be alerted by fax. Patients automatically received a tollfree 24-hour suicide hotline number and could transfer there immediately. Finally, patients were informed that their caregiver (if applicable) would receive automated information by phone about assisting the patient in getting help.

Poor Medication Adherence. Alerts were generated if the patient reported “rarely or never” taking their antidepressant as prescribed, or that they had considered reducing their medication or stopping medications due to side effects in the past week.

Increase in Depressive Symptom Severity. Alerts were generated if patients’ PHQ-9 scores increased by at least 5 points since the preceding assessment, or increased from lower than 15 to 15 or higher.

Data Collection and Analysis

We analyzed data at the patient-week level (ie, 1 record for each week an assessment was attempted). At baseline we collected data on patients’ sociodemographic and clinical characteristics including PHQ-8³⁷ (the PHQ-9 without the suicidality item), the number of comorbid medical conditions (hypertension, cardiovascular disease, hyperlipidemia, stroke, arthritis, chronic lung disease, and low back pain), scores on the Medical Outcomes Study Short Form 12 (SF-12),³⁸ hospitalizations, and missed outpatient visits in the prior year.

We examined differences in characteristics of patients who did versus did not enroll with a caregiver. We used 2 multivariable logistic regression models to examine patients’ probability of completing assessments as a function of their demographic characteristics, measures of baseline vulnerability (missed appointments, hospitalizations, depressive symptoms, SF-12 physical and mental functioning scores), and weeks of follow-up. With the exception of patients’ baseline depression scores, coefficients in the full model with P values greater than .15 were dropped from the final model. Based on model coefficients we used the prgen command in Stata version 11.2³⁹ to plot the probability of assessment completion by patients’ baseline depressive symptom scores. We fit a second logistic model to identify whether assessment completion rates were affected by whether the patient elected to receive monthly versus weekly calls. Because reporting 3 successive PHQ-9 scores lower than 10 was a criterion for monthly calls, patients who did not complete calls regularly would be less likely to be represented in the monthly call data. We minimized that bias by including only patient call-weeks for which the prior assessment was successfully completed. Both models corrected coefficient standard errors for clustering of personweeks within patients.

Finally, we examined the frequency of clinical alerts per 100 patient-weeks of program participation. We used logistic regression to identify systematic variation in the probability of each alert type across groups of patients defined by their baseline depressive symptom severity and controlling for sociodemographic covariates. In the subset of patients enrolled from the university-affiliated clinics, we reviewed medical records to characterize clinicians’ response to alerts generated based on reports of suicidal ideation.

RESULTSPatient Characteristics

Table 1

A total of 1330 potentially eligible patients were contacted, of which 387 primary care patients from 13 university-based (81%) and community-based (19%) clinics were enrolled. Participants tended to be female (80%) and white (90%), to have more than a high school education (80%), and to be married (62%; ). There were a broad range of ages, with 31% being 21 to 45 years old and 12% being 66 years or older. At baseline, 52% reported at least moderate depressive symptoms and 26% reported at least moderately severe symptoms. Two-thirds of patients (66%) reported missing 1 or more primary care appointments in the prior year, and 24% reported being hospitalized in the prior year. Patients reported a mean of 2.1 comorbid medical conditions, with the most prevalent being arthritis (48%), hypertension (47%), back pain (39%), and chronic lung disease (31%).

Most patients enrolled with a caregiver (59%), and these patients had significantly lower incomes, more severe baseline depressive symptoms, a greater likelihood of having been hospitalized in the prior year, and more comorbid chronic conditions(Table 1). Compared with patients who did not enroll with a caregiver, those with a caregiver were more likely to have hypertension (52% vs 40%), chronic lung disease (34% vs 26%), and chronic back pain (42% vs 34%) (each P <.05).

Assessment Completion Rates

Table 2

Patients participated for a median of 25 weeks (interquartile range: 21 weeks to 48 weeks), yielding 12,042 patientweeks of follow-up (). Eleven percent dropped out in the first 6 months. Attrition was unrelated to baseline depressive symptoms and other clinical and sociodemographic characteristics. Interactive voice response assessments were attempted during 7912 call-weeks, with the remaining 4130 being protocol-driven “gap weeks” due to patients opting for monthly assessments or to temporary call suspension (eg, vacation). Patients completed assessments during 68% of the attempted call-weeks, yielding 5360 assessments.

Out of the 5360 completed assessments, 582 (10.9%) were assessments in which the patient was given the opportunity to switch from weekly to monthly calls because of 3 prior assessments indicating well-controlled depressive symptoms. During most of those calls (364/582; 63%), the patient opted to continue with weekly calling. In 96 monthly assessments, the patient’s PHQ-9 depression scores were 10 or higher, and as a result, the patient was given the option of returning to weekly calling. Those patients chose to return to weekly calls 41% of the time (39/96). Overall, 19.2% of attempted callweeks followed month-long gaps.

As shown in model 1 in Table 3, the odds of a completed assessment were greater when patients had a caregiver, were older, were married, and had a greater number of comorbid conditions. Completion rates did not vary significantly according to patients’ baseline SF-12 physical functioning scores, sex, income, educational attainment, or race (all P >.15), and these variables were dropped from the final model. The likelihood of assessment completion decreased over time. Importantly, completion rates did not vary significantly according to patients’ baseline depression symptom severity (Table 3 and Figure 1).

Considering only call-weeks that followed a successfully completed assessment (Table 3, model 2), the odds of call completion were lower when eligible patients opted for monthly instead of weekly calls. This effect was not attributable to patients’ depressive symptom severity at baseline or during the preceding completed assessment.

Clinician Alerts

Alerts were triggered at a rate of 4.9 per 100 person-weeks of participation (Table 2). More than half were triggered by poor antidepressant adherence (n = 349 alerts for 133 patients; ie, 2.8 per 100 person-weeks). Another 2.2 alerts per 100 personweeks (n = 272 alerts for 137 patients) were triggered by increased depressive symptom severity, while a small number (n = 29 alerts for 19 patients; ie, 0.2% alerts per 100 person-weeks) were triggered by reports of suicidal ideation. As expected, clinical alerts triggered by poor medication adherence and suicidal ideation increased monotonically with baseline depressive symptom severity (Figure 2). In contrast, alerts triggered by increased depressive symptom severity had a curvilinear association with baseline symptom severity, with a peak in frequency when the patient’s baseline score was approximately 15.

We examined clinicians’ responses to 15 suicidal ideation alerts through chart review for patients who received care from university-affiliated practices. Clinician follow-up was documented for 13 (87%) of these incidents. Follow-up occurred on the same day for 8 incidents and within 3 days for 11 incidents. For 7 of these incidents, medical records documented that the patient reported on follow-up that he or she indicated suicidal ideation accidentally. While the samples are too small

for statistical comparisons, these 7 patients did not differ from the other 6 patients who reported suicidal thoughts in terms of either the internal consistency or test-retest correlation of their PHQ-9 responses. For all 13 patients who received clinical follow-up, the medical record indicated that the patient underwent review of symptoms, adherence, and psychosocial stressors; 5 notes documented review of medications and/or appointments with mental health providers. No reviewed records indicated that emergency intervention was required or that a suicidal attempt occurred.

DISCUSSION

Prior studies describe the use of IVR assessment and selfmanagement support calls to monitor symptoms among patients with mental health and substance abuse disorders.^40-43 However, most of these studies had small samples and relatively brief follow-up. Here, we demonstrated that an automated IVR service is a feasible long-term strategy to improve between-visit monitoring and self-management support of depressed primary care patients. Several important findings merit emphasis.

Given the option, most patients participated with a caregiver who received updates on the patient’s status with targeted self-management support advice, suggesting that health systems should strive to balance privacy considerations with patients’ preferences to engage social network members. Although stronger social ties are associated with better health status,^{25,29,30,44-46} at baseline patients participating with a caregiver had more severe depressive symptoms, were more likely to have been hospitalized in the prior year, and reported more comorbid diagnoses compared with patients participating alone. This suggests that the social support option was most likely to be used by patients needing it the most. Despite their worse health status, patients participating with a caregiver were significantly more likely to complete IVR calls.

Call completion was unrelated to sociodemographic characteristics, suggesting that engagement is possible even among vulnerable patients. Call completion was also unrelated to baseline depressive symptom severity, even though patients with more severe depression struggle with low energy, social withdrawal, and low motivation. Call completion was at least as high among older patients and higher among patients with multiple chronic health conditions. These encouraging findings suggest that IVR may be useful for improving access to monitoring and self-care support among patients with depression and complex chronic conditions.

As shown in Table 3, the likelihood of call completion substantially decreased when stabilized patients opted to change from weekly to monthly calls. This decrease was unassociated with recent depressive symptom severity. Future studies should weigh the respondent burden of weekly monitoring for relatively healthy patients against the possibility that less frequent monitoring may lead to reduced engagement.

Although clinicians are understandably concerned about the additional workload generated from between-visit automated symptom monitoring, we found that only a modest number of clinical alerts were generated using the current algorithms. Importantly, reports of suicidal ideation were rare, and other alerts could be triaged by a nursing assistant. Alerts related to worsening depressive symptoms (Figure 2) were not necessarily distributed across patients in a predictable way according to baseline depression scores. Ongoing analyses of alert frequency and distribution remain important in order to clarify implications of these programs for clinical follow-up. Among patients reporting suicidal ideation where follow-up was documented, half indicated mistakenly reporting thoughts of self-harm. If true, this suggests that follow-up for erroneous suicide risks occurred at roughly 1 patient per 1000 patient-weeks of follow-up (ie, half the rate of suicide reports in Table 2). However, multiple studies have shown that IVR-reported psychiatric problems are as reliable as in-person reports,^47-50 and analyses of data from this study suggest that patients who reported erroneous suicidal ideation had PHQ scores at least as reliable as those of other patients.

This study has several of those limitations. While the sample was heterogeneous in terms of age, depressive symptom severity, and primary care setting, the majority were white and female with at least some post high school education. Most potentially eligible patients (71%) did not enroll in this program, and unfortunately, due to human subjects restrictions we were not able to collect information allowing us to compare those patients’ sociodemographic and clinical characteristics with those of enrollees. Future work on understanding the barriers to enrollment will be important, and a portfolio of options for improving depression self-care (eg, including peer support⁵¹) will likely be needed to meet other patients’ needs. Alert rates shown in Figure 2 suggest that the total number of alerts generated will depend on the distribution of patients’ depression severity on enrollment. Thus, the volume could be higher if patients with more severe depression were targeted. Other than the analyses of suicidal ideation alerts, we did not measure the impact on primary or specialty care providers of telephone or face-to-face encounters (resulting from the IVR alerts). Finally, patients had high rates of comorbid medical conditions, and the current IVR patient messages did not mention the importance of managing depression for health outcomes in general. Nevertheless, it would be important in future studies to determine the impact of these services on patients’ overall health and functioning.

In conclusion, we found that implementing IVR assessment and self-management support for depressed primary care patients was feasible, was well accepted by patients and caregivers, and did not generate an unmanageable clinician burden. Patients’ engagement was greater when they participated with a caregiver who received ongoing updates about their mood and functioning along with advice on how to support their efforts at depression self-management. Automated depression monitoring and self-management support programs may help fill the gap between what some patients need or want and what resource-constrained health systems can realistically provide.Author Affiliations: From Center for Clinical Management Research (JDP, NSM, DS, SJB), VA Ann Arbor Healthcare System, Ann Arbor, MI; Department of Internal Medicine (JDP, DP, CS, NSM, DS, SJB), University of Michigan Medical School, Ann Arbor, MI; Department of Family Medicine (JEA), University of Michigan Medical School, Ann Arbor, MI; Northwest HSR&D Center of Excellence in Outcomes Among Older Adults (RT), VA Puget Sound Health Care System, Seattle, WA; Department of Health Services (RT), University of Washington School of Public Health, Seattle, WA.

Funding Source: Funding was provided by the University of Michigan Health System Faculty Group Practice, Blue Cross and Blue Shield of Michigan, and grant R01 MH096699-01 from the National Institute of Mental Health. Funding for Dr Piette and Dr Trivedi was from their VA Career Scientist and Career Development awards. Other financial support came from grant number P30DK092926 from the National Institute of Diabetes and Sigestive Kidney Diseases.

Author Disclosures: The authors (JDP, JEA, RT, DP, CS, NSM, DS, SJB) report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article. Authorship Information: Concept and design (JDP); acquisition of data (JDP, DP, NSM, DS); analysis and interpretation of data (JDP, JEA, CS); drafting of the manuscript (JDP, JEA, RT, SJB); critical revision of the manuscript for important intellectual content (JDP, JEA, RT, CS, NSM, DS, SJB); statistical analysis (JDP); provision of study materials or patients (JDP, DP, CS); obtaining funding (JDP, SJB); administrative, technical, or logistic support (DP, NSM, DS, SJB); and supervision (JDP).

Address correspondence to: John D. Piette, MSc, PhD, Center for Clinical Management Research HSR&D Center of Excellence, VA Ann ArboHealthcare System, PO Box 130170, Ann Arbor, MI 48113-0170. E-mail: jpiette@umich.edu.1. Kessler RC, Chiu WT, Demler O, Merikangas KR, Walters EE. Prevalence, severity, and comorbidity of 12-month DSM-IV disorders in the National Comorbidity Survey Replication. Arch Gen Psychiatry. 2005;62(6):617-627.

2. Mathers C, Fat D, Boerma J. The Global Burden of Disease: 2004 Update. Geneva, Switzerland: World Health Organization; 2008.

3. Katon W, Ciechanowski P. Impact of major depression on chronic medical illness. J Psychosom Res. 2002;53(4):859-863. 4. Cuijpers P, Smit F. Excess mortality in depression: a meta-analysis of community studies. J Affect Disord. 2002;72(3):227-236.

5. Simon GE. Social and economic burden of mood disorders. Biol Psychiatry. 2003;54(3):208-215.

6. Katon W, Schulberg H. Epidemiology of depression in primary care. Gen Hosp Psychiatry. 1992;14(4):237-247.

7. Katon WJ. Epidemiology and treatment of depression in patients with chronic medical illness. Dialogues Clin Neurosci. 2011;13(1):7-23.

8. Williams JW Jr, Rost K, Dietrich AJ, Ciotti MC, Zyzanski SJ, Cornell J. Primary care physicians’ approach to depressive disorders. Effects of physician specialty and practice structure. Arch Fam Med. 1999; 8(1):58-67.

9. Scogin FR, Hanson A, Welsh D. Self-administered treatment in stepped-care models of depression treatment. J Clin Psychol. 2003;59(3): 341-349.

10. Aikens JE, Nease JD Jr, Nau DP, Klinkman MS, Schwenk TL. Adherence to maintenance-phase antidepressant medication as a function of patient beliefs about medication. Ann Fam Med. 2005;3(1):23-30.

11. Pignone MP, Gaynes BN, Rushton JL, et al. Screening for depression in adults: a summary of the evidence for the U.S. Preventive Services Task Force. Ann Intern Med. 2002;136(10):765-776.

12. Ballenger JC, Davidson JR, Lecrubier Y, et al. Consensus statement on the primary care management of depression from the International Consensus Group on Depression and Anxiety. J Clin Psychiatry. 1999;60(suppl 7):54-61.

13. Williams JW. Feedback to physicians plus telephone care management improved outcomes in primary care patients with depression. Evid Based Med. 2000;5(5):157.

14. Simon GE, Ludman EJ, Tutty S, Operskalski B, Von Korff M. Telephone psychotherapy and telephone care management for primary care patients starting antidepressant treatment: a randomized controlled trial. JAMA. 2004;292(8):935-942.

15. Yarnall KS, Pollak KI, Østbye T, Krause KM, Michener JL. Primary care: is there enough time for prevention? Am J Public Health. 2003; 93(4):635-641.

16. Konrad TR, Link CL, Shackelton RJ, et al. It’s about time: physicians’ perceptions of time constraints in primary care medical practice in three national healthcare systems. Med Care. 2010;48(2):95-100.

17. Piette JD, Rosland AM, Marinec NS, Striplin D, Bernstein SJ, Silveira MJ. Engagement in automated patient monitoring and selfmanagement support calls: experience with a thousand chronically ill patients. Med Care. 2013;51(3):216-223.

18. Ben-Zeev D, Davis KE, Kaiser S, Krzsos I, Drake RE. Mobile technologies among people with serious mental illness: opportunities for future services. Adm Policy Ment Health. 2013;40(4):340-343.

19. Agyapong VI, Ahern S, McLoughlin DM, Farren CK. Supportive text messaging for depression and comorbid alcohol use disorder: singleblind randomised trial. J Affect Disord. 2012;141(2-3):168-176.

20. Ben-Zeev D, Young MA, Depp CA. Real-time predictors of suicidal ideation: mobile assessment of hospitalized depressed patients. Psychiatr Res. 2012;197(1-2):55-59.

21. Kimhy D, Myin-Germeys I, Palmier-Claus J, Swendsen J. Mobile assessment guide for research in schizophrenia and severe mental disorders. Schizophr Bull. 2012;38(3):386-395.

22. Kobak KA, Greist JH, Jefferson JW, Mundt JC, Katzelnick DJ. Computerized assessment of depression and anxiety over the telephone using interactive voice response. MD Comput. 1999;16(3):64-68.

23. Schwartz S. What works. IVR system helps diabetes center triple patient load; increases care level. Health Manag Technol. 1998;19(9):41.

24. Friedman RH, Kazis LE, Jette A, et al. A telecommunications system for monitoring and counseling patients with hypertension: impact onmedication adherence and blood pressure control. Am J Hypertens. 1996;9(4, pt 1):285-292.

25. Harris T, Brown GW, Robinson R. Befriending as an intervention for chronic depression among women in an inner city, 1: randomised controlled trial. Br J Psychiatry. 1999;174:219-224.

26. Stice E, Ragan J, Randall P. Prospective relations between social support and depression: differential direction of effects for parent and peer support? J Abnorm Psychol. 2004;113(1):155-159.

27. Gladstone GL, Parker GB, Malhi GS, Wilhelm KA. Feeling unsupported? an investigation of depressed patients’ perceptions. J Affect Disord. 2007;103(1-3):147-154.

28. Burcusa SL, Iacono WG. Risk for recurrence in depression. Clin Psychol Rev. 2007;27(8):959-985.

29. Lynch TR, Mendelson T, Robins CJ, et al. Perceived social support among depressed elderly, middle-aged, and young-adult samples: cross-sectional and longitudinal analyses. J Affect Disord. 1999;55(2-3): 159-170.

30. Strine TW, Kroenke K, Dhingra S, et al. The associations between depression, health-related quality of life, social support, life satisfaction, and disability in community-dwelling US adults. J Nerv Ment Dis. 2009;197(1):61-64.

31. Talley RC, Crews JE. Framing the public health of caregiving. Am J Public Health. 2007;97(2):224-228.

32. Martire LM, Lustig AP, Schulz R, Miller GE, Helgeson VS. Is it beneficial to involve a family member? a meta-analysis of psychosocial interventions for chronic illness. Health Psychol. 2004;23(6):599-611.

33. Ekwall A, Sivberg B, Hallberg IR. Dimensions of informal care and quality of life among elderly family caregivers. Scand J Caring Sci. 2004;18(3):239-248.

34. Armour TA, Norris SL, Jack L Jr, Zhang X, Fisher L. The effectiveness of family interventions in people with diabetes mellitus: a systematic review. Diabet Med. 2005;22(10):1295-1305.

35. Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med. 2001;16(9):606-613.

36. Norbeck JS. Norbeck Social Support Questionnaire. Questionnaires/Research Instruments. http://nursing.ucsf.edu/jane-norbeckquestionnaires-research-instruments. Published 1995. Accessed September 18, 2013.

37. Kroenke K, Strine TW, Spitzer RL, Williams JB, Berry JT, Mokdad AH. The PHQ-8 as a measure of current depression in the general population. J Affect Disord. 2009;114(1-3):163-173.

38. Jenkinson C, Layte R, Jenkinson D, et al. A shorter form health survey: can the SF-12 replicate results from the SF-36 in longitudinal studies? J Public Health Med. 1997;19(2):179-186.

39. StataCorp. Stata Statistical Software [computer program]. Release 11. College Station, TX: StataCorp LP; 2009.

40. Aharonovich E, Hatzenbuehler ML, Johnston B, et al. A low-cost, sustainable intervention for drinking reduction in the HIV primary care setting. AIDS Care. 2006;18(6):561-568.

41. Osgood-Hynes DJ, Greist JH, Marks IM, et al. Self-administered psychotherapy for depression using a telephone-accessed computer system plus booklets: an open U.S.-U.K. study. J Clin Psychiatry. 1998;59(7):358-365.

42. Kim H, Bracha Y, Tipnis A. Automated depression screening in disadvantaged pregnant women in an urban obstetric clinic. Arch Womens Ment Health. 2007;10(4):163-169.

43. Brodey BB, Rosen CS, Brodey IS, Sheetz B, Unutzer J. Reliability and acceptability of automated telephone surveys among Spanishand English-speaking mental health services recipients. Ment Health Serv Res. 2005;7(3):181-184.

44. Spiegel D, Bloom JR, Kraemer HC, Gottheil E. Effect of psychosocial treatment on survival of patients with metastatic breast cancer. Lancet. 1989;2(8668):888-891.

45. Smith RT, Hämäläinen H, Kallio V, et al. Resource support and heart patient recovery. Int J Rehabil Res. 1997;20(1):11-28.

46. Gallant MP. The influence of social support on chronic illness selfmanagement:a review and directions for research. Health Educ Behav. 2003;30(2):170-195.

47. Kobak KA, Taylor LH, Dottl SL, et al. A computer-administered telephone interview to identify mental disorders. JAMA. 1997;278(11): 905-910.

48. Moore HK, Mundt JC, Modell JG, et al. An examination of 26,168 Hamilton Depression Rating Scale scores administered via interactive voice response across 17 randomized clinical trials. J Clin Psychopharmacol. 2006;26(3):321-324.

49. Mundt JC, Katzelnick DJ, Kennedy SH, Eisfeld BS, Bouffard BB, Greist JH. Validation of an IVRS version of the MADRS. J Psychiatr Res. 2006;40(3):243-246.

50. Turvey CL, Willyard D, Hickman DH, Klein DM, Kukoyi O. Telehealth screen for depression in a chronic illness care management program. Telemed J E Health. 2007;13(1):51-56.

51. Pfeiffer PN, Heisler M, Piette JD, Rogers MA, Valenstein M. Efficacy of peer support interventions for depression: a meta-analysis. Gen Hosp Psychiatry. 2011;33(1):29-36.

Download PDF: Depression Self-Management Assistance Using Automated Telephonic Assessments and Social Support