Long Term Effect of Depression Care Management on Mortality
Long Term Effect of Depression Care Management on Mortality
PROSPECT was conducted in 20 primary care practices located in greater New York City, Philadelphia, and Pittsburgh from May 1999 to August 2001. After pairing by urban location, academic affiliation, size, and population type, practices were randomized by coin flip to the intervention condition or to usual care (cluster randomization by practice). Patients were recruited from an age stratified (60-74, ≥75 years) random sample of people with upcoming appointments. Research associates confirmed eligibility (age ≥60 years, mini-mental state examination score >17, English speaking) of consenting patients and screened them for depression with the Centers for Epidemiologic Studies depression scale (CES-D). All patients with a CES-D score above 20 were invited to enroll, as was a 5% sample with lower scores. Research associates met patients at the practice, obtained signed informed consent, and administered a baseline interview.
Major depression was diagnosed on the basis of standard criteria contained in the Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV). Clinically significant minor depression was defined by DSM-IV criteria for minor depression modified by requiring four depressive symptoms, Hamilton depression rating scale score of 10 or more, and duration of at least four weeks. We did a structured clinical interview for axis I DSM-IV disorders (SCID) assessment of all participants.
We obtained baseline information on age, sex, marital status, self reported ethnicity, educational attainment, and smoking based on tobacco use within six months. Patients self reported medical conditions. The 24 item Hamilton depression rating scale measured severity of depression, and the scale for suicidal ideation indicated the presence of suicidal ideation.
Practices randomized to usual care received educational sessions for primary care physicians and notification of the depression status of patients. Physicians received no specific recommendations regarding individual patients (except for psychiatric emergencies). Practices randomized to the intervention condition received educational sessions for primary care physicians, education of patients' families, and a depression care manager who worked within the practice. The care manager worked with the primary care physician to recommend treatment according to standard guidelines. Care managers had psychiatric back-up including on-demand consultation, weekly supervision by psychiatrist investigators, and monthly interpersonal therapy cross site supervision. The 15 care managers included social workers, nurses, and psychologists who interacted with patients in person or by telephone at scheduled intervals and as necessary. Care managers monitored symptoms, adverse effects of drugs, and adherence to treatment.
The PROSPECT algorithm provided guidelines to care managers for target and maximum daily doses of antidepressants. After six weeks, the target dosage was maintained if the patient showed a significant improvement (>50% reduction in Hamilton depression rating scale score) but was increased if the improvement was partial (30-50% reduction in score). For patients who had not responded at 12 weeks, the care manager followed guidelines for switching antidepressants. Interpersonal psychotherapy for depression could be used alone or as an augmentation strategy depending on whether the patient tolerated the antidepressant and the presence or absence of a partial response. In both study arms, physicians were informed by letter if patients reported any suicidal ideation and were immediately informed when patients were identified to be at high risk of suicide, according to pre-specified guidelines. Other sources detail the PROSPECT treatment algorithm and implementation, including the role of the care manager, the strategy for drug treatment, management of suicidal ideation, and types and proportions of treatment received over time by patients in practices randomized to the intervention condition or to usual care.
Vital status was based on follow-up of participants through the National Center for Health Statistics national death index (NDI Plus). Each participant gave written consent, including permission to obtain information from death certificates.
We compared baseline characteristics of patients across baseline depression status (major depression or minor depression versus no depression) stratified by groups defined by practice randomization assignment (intervention condition or usual care), by using linear and logistic regression with random effects to account for clustering of patients by practice. Our primary strategy was to compare people with depression and people without depression within intervention strata, followed by comparison of people with depression across intervention strata. On the basis of the age, sex, and ethnic composition of the PROSPECT sample and using mortality rates for the US population, we estimated that 300 to 500 deaths would occur among the 1226 patients, a number of deaths large enough to detect a clinically meaningful signal accounting for the study design.
We modeled survival time with Cox proportional hazards regression, adjusting standard errors for within practice clustering. Final models included terms for characteristics identified by their association (P<0.05) with time to death: baseline age, sex, education, marital status, smoking status, cardiovascular disease, stroke, diabetes, cancer, cognition, and suicidal ideation. We assessed the proportional hazard assumption by including time dependent terms in the unadjusted model and measuring the global effect. We used SAS version 9.1 and Stata version 12.0 for these analyses. We plotted survival time by using the method of Kaplan and Meier. We set α at 0.05 to denote statistical significance, recognizing that tests of statistical significance are approximations that serve as aids to interpretation and inference.
Methods
Study Sample
PROSPECT was conducted in 20 primary care practices located in greater New York City, Philadelphia, and Pittsburgh from May 1999 to August 2001. After pairing by urban location, academic affiliation, size, and population type, practices were randomized by coin flip to the intervention condition or to usual care (cluster randomization by practice). Patients were recruited from an age stratified (60-74, ≥75 years) random sample of people with upcoming appointments. Research associates confirmed eligibility (age ≥60 years, mini-mental state examination score >17, English speaking) of consenting patients and screened them for depression with the Centers for Epidemiologic Studies depression scale (CES-D). All patients with a CES-D score above 20 were invited to enroll, as was a 5% sample with lower scores. Research associates met patients at the practice, obtained signed informed consent, and administered a baseline interview.
Assessment of Depression
Major depression was diagnosed on the basis of standard criteria contained in the Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV). Clinically significant minor depression was defined by DSM-IV criteria for minor depression modified by requiring four depressive symptoms, Hamilton depression rating scale score of 10 or more, and duration of at least four weeks. We did a structured clinical interview for axis I DSM-IV disorders (SCID) assessment of all participants.
Assessment of Patients' Characteristics
We obtained baseline information on age, sex, marital status, self reported ethnicity, educational attainment, and smoking based on tobacco use within six months. Patients self reported medical conditions. The 24 item Hamilton depression rating scale measured severity of depression, and the scale for suicidal ideation indicated the presence of suicidal ideation.
Usual Care and Intervention Conditions
Practices randomized to usual care received educational sessions for primary care physicians and notification of the depression status of patients. Physicians received no specific recommendations regarding individual patients (except for psychiatric emergencies). Practices randomized to the intervention condition received educational sessions for primary care physicians, education of patients' families, and a depression care manager who worked within the practice. The care manager worked with the primary care physician to recommend treatment according to standard guidelines. Care managers had psychiatric back-up including on-demand consultation, weekly supervision by psychiatrist investigators, and monthly interpersonal therapy cross site supervision. The 15 care managers included social workers, nurses, and psychologists who interacted with patients in person or by telephone at scheduled intervals and as necessary. Care managers monitored symptoms, adverse effects of drugs, and adherence to treatment.
The PROSPECT algorithm provided guidelines to care managers for target and maximum daily doses of antidepressants. After six weeks, the target dosage was maintained if the patient showed a significant improvement (>50% reduction in Hamilton depression rating scale score) but was increased if the improvement was partial (30-50% reduction in score). For patients who had not responded at 12 weeks, the care manager followed guidelines for switching antidepressants. Interpersonal psychotherapy for depression could be used alone or as an augmentation strategy depending on whether the patient tolerated the antidepressant and the presence or absence of a partial response. In both study arms, physicians were informed by letter if patients reported any suicidal ideation and were immediately informed when patients were identified to be at high risk of suicide, according to pre-specified guidelines. Other sources detail the PROSPECT treatment algorithm and implementation, including the role of the care manager, the strategy for drug treatment, management of suicidal ideation, and types and proportions of treatment received over time by patients in practices randomized to the intervention condition or to usual care.
Ascertainment of Vital Status
Vital status was based on follow-up of participants through the National Center for Health Statistics national death index (NDI Plus). Each participant gave written consent, including permission to obtain information from death certificates.
Analysis Strategy
We compared baseline characteristics of patients across baseline depression status (major depression or minor depression versus no depression) stratified by groups defined by practice randomization assignment (intervention condition or usual care), by using linear and logistic regression with random effects to account for clustering of patients by practice. Our primary strategy was to compare people with depression and people without depression within intervention strata, followed by comparison of people with depression across intervention strata. On the basis of the age, sex, and ethnic composition of the PROSPECT sample and using mortality rates for the US population, we estimated that 300 to 500 deaths would occur among the 1226 patients, a number of deaths large enough to detect a clinically meaningful signal accounting for the study design.
We modeled survival time with Cox proportional hazards regression, adjusting standard errors for within practice clustering. Final models included terms for characteristics identified by their association (P<0.05) with time to death: baseline age, sex, education, marital status, smoking status, cardiovascular disease, stroke, diabetes, cancer, cognition, and suicidal ideation. We assessed the proportional hazard assumption by including time dependent terms in the unadjusted model and measuring the global effect. We used SAS version 9.1 and Stata version 12.0 for these analyses. We plotted survival time by using the method of Kaplan and Meier. We set α at 0.05 to denote statistical significance, recognizing that tests of statistical significance are approximations that serve as aids to interpretation and inference.
Source...