HHS Public Access Author manuscript Author Manuscript
Psychiatr Serv. Author manuscript; available in PMC 2017 August 15. Published in final edited form as: Psychiatr Serv. 2016 June 01; 67(6): 636–641. doi:10.1176/appi.ps.201400465.
Screening and diagnosis of depression in adolescents in a large HMO R. Eric Lewandowski, Ph.D., Department of Child and Adolescent Psychiatry, New York University, New York
Author Manuscript
Briannon O’Connor, Ph.D., Department of Child and Adolescent Psychiatry, New York University, New York Andrew Bertagnolli, Ph.D., Care Management Institute, Kaiser Permanente, Oakland, California Aldo Tinoco, M.D., National Committee for Quality Assurance, Washington, D.C Arne Beck, Ph.D., Institute for Health Research, Kaiser Permanente of Colorado, Denver William Gardner, Ph.D., Research Institute of Children’s Hospital of Eastern Ontario, Ottawa, Ontario, Canada Department of Pediatrics, Ohio State University, Columbus
Author Manuscript
Christine Jelinek-Berents, M.D., Department of Pediatrics, Kaiser Permanente of Colorado, Denver Doug Newton, M.D., Care Management Institute and the, Department of Child and Adolescent Psychiatry, Kaiser Permanente of Colorado, Denver Kris Wain, M.S., Institute for Health Research, Kaiser Permanente of Colorado, Denver Jennifer Boggs, M.S., Institute for Health Research, Kaiser Permanente of Colorado, Denver
Author Manuscript
Nancy Brace, R.N., Institute for Health Research, Kaiser Permanente of Colorado, Denver Patricia De Sa, Ph.D., Care Management Institute, Kaiser Permanente, Oakland, California Sarah H. Scholle, Dr.P.H., National Committee for Quality Assurance, Washington, D.C Kimberly Hoagwood, Ph.D., and
Disclosures The authors report no conflicts of interest or other disclosures.
Lewandowski et al.
Page 2
Department of Child and Adolescent Psychiatry, New York University, New York
Author Manuscript
Sarah M. Horwitz, Ph.D. Department of Child and Adolescent Psychiatry, New York University, New York
Abstract OBJECTIVE—The aim of this analysis was to determine changing patterns of depression screening and diagnosis over three years in primary and specialty mental health care in a large health maintenance organization (HMO) as part of a quality measure development project.
Author Manuscript
METHODS—Two series of aggregate data spanning three years (2010–12) were gathered from the electronic health record of the HMO summarizing unique adolescents age 12–21 (N=44,342) who had visits in primary and mental health care. Chi-square tests assessed the significance of changing frequency and departmental location of PHQ-9 administration, incident depression symptoms, and depression diagnoses. RESULTS—There was a significant increase in PHQ-9 use from 2010 to 2012, predominantly in primary care, consistent with internally generated organizational recommendations to increase screening with the PHQ-9. The increase in PHQ-9 use led to an increase in depression diagnoses in primary care and shift in the location of some diagnoses from specialty mental health care to primary care. The increase in PHQ-9 use was also linked to an increase in the proportion of positive PHQ-9 results not leading to formal depression diagnoses.
Author Manuscript
CONCLUSIONS—The rate of depression screening in primary care increased over the study period. This increase corresponded to an increase in the number of depression diagnoses made in primary care and a shift in the location of depression diagnoses made from the mental health department to primary care. The frequency of positive PHQ-9 administrations not associated with depression diagnoses also increased. Depression affects 12 to 25% of adolescents,1,2 and is associated with a range of negative academic, social, and health outcomes.3–6 Despite the high burden of depression and the availability of effective treatments, approximately 60% to 80% of affected adolescents do not receive appropriate care.7–9 Pediatric primary care is an important site for the identification of depression in adolescents, the critical first step in connecting youth to treatment.10,11 However, depression remains poorly identified in this setting.12–15 Professional organizations and expert consensus in pediatrics and psychiatry have recommended routine depression screening of adolescents to improve case identification, though available evidence suggests that screening coverage is very low.16–18
Author Manuscript
Quality measurement is an increasingly prominent approach for improving care, whereby providers and organizations are incentivized to adhere to essential healthcare practices.19–22 Depression screening in primary care has been recommended by the United States Preventive Services Task Force (USPSTF) and other major practice guidelines, and has featured prominently in efforts to improve depression care,12,15 including in national initiatives to develop quality measures. For example, the Center for Medicare and Medicaid Services (CMS) included a quality measure of depression screening and follow up for adolescents and adults in the Electronic Health Record Incentive program for “meaningful
Psychiatr Serv. Author manuscript; available in PMC 2017 August 15.
Lewandowski et al.
Page 3
Author Manuscript
use” of electronic health records (EHR). Similarly, the National Collaborative for Innovation in Quality (NCINQ), one of seven centers of excellence funded under the Agency for Health Research and Quality and CMS Pediatric Quality Measures Program developed a suite of potential quality measures organized around a care pathway for managing adolescent depression in primary care in which the first step, and first potential quality measure, is screening for depression.22
Author Manuscript
METHOD
Author Manuscript
As the next step in quality measure development, NCINQ partnered with a large HMO to study changing patterns of depression screening over time. This study also assessed depression diagnoses following screening, which corresponds to the second measure in the NCINQ care pathway, and directly addresses the specific research need highlighted by USPSTF to link screening to improved identification of depression cases.16 Large scale, naturalistic data on screening and diagnosis will help to identify gaps in essential care practices related to depression case identification, and provide an indication of the fit and possible value of the depression quality measures in the context of current practice.
Measures
Setting and data sources Data for adolescents meeting inclusion criteria were abstracted from the EHR of a large HMO. The identity of the HMO has been masked per agreement with the organization. The Chesapeake Institutional Review Board determined that this study was not human subjects research since only de-identified, aggregated data were collected. Starting in 2011, the HMO began to implement the adolescent version of the Patient Health Questionnaire-9 (PHQ-9) as a screening tool in primary care.23, 24 Previously, the PHQ-9 was used primarily in the mental health department for diagnostic support and monitoring of patients with known diagnosis of depression. Computer programmed extraction of aggregate data for this study was conducted from May 29–31, 2013.
Author Manuscript
The Patient Health Questionnaire-9-Modified form—The PHQ-9 is a 9-item selfreport questionnaire assessing depression symptoms and severity that has been validated its original form with adolescents.23,24 Items based on the DSM-IV criteria for depression are rated on a 4-point scale ranging from 0 (“not at all”) to 3 (“nearly every day”). The PHQ-9Modified includes minimal adjustments to the original PHQ-9 to incorporate characteristics of depression in adolescents and age-appropriate language. Specifically, the PHQ-9Modified includes irritability in the item assessing depressed mood, and weight loss to the item assessing appetite. No psychometric data is available for the PHQ-9-Modified, but because it is identical to the PHQ-9, besides the described adjustments, the developers indicate that using PHQ-9 cutoffs is appropriate. Throughout this paper “PHQ-9” is used to indicate the PHQ-9-Modified. Consistent with research literature and practices within the HMO, a PHQ-9 score of 11 or greater indicated a positive screen.24 In this study, a positive PHQ-9 score that was preceded by a period of 6 months with no record of depression diagnosis or antidepressant medication ordered, was termed an “incident positive PHQ-9” (distinguishing it from repeated PHQ-9 administrations with the same patient to track
Psychiatr Serv. Author manuscript; available in PMC 2017 August 15.
Lewandowski et al.
Page 4
Author Manuscript
symptoms in a prevalent case) and taken to represent the identification of new depression symptoms. The 6-month depression-free period preceding a positive PHQ-9 began 14 days prior to the PHQ-9 chart entry date to allow for institutional lag time between PHQ-9 administration and upload into the EHR. Depression diagnosis—Depression diagnoses in the current study included ICD-9 codes for Major Depressive Affective Disorder, Depressive-type Psychosis, Depressive DisorderNOS, Adjustment Disorder with Depressed Mood, with Mixed Anxiety and Depressed Mood, with Mixed Disturbance of Emotions and Conduct.25 A new depression diagnosis was defined as one preceded by a period of 6 months with no depression diagnosis or antidepressant medication order within the EHR. Study Group
Author Manuscript
For three years (2010, 2011, and 2012), adolescents were included in the study sample if they were 12 to 21 years of age on December 31st of the previous year and if their EHR contained documentation of at least one face-to-face visit with a provider. Adolescents may have had multiple visits and multiple PHQ-9 results in a given calendar year, but were counted only once. Adolescents receiving more than one PHQ-9 in a given calendar year were counted in the department where the first PHQ-9 of the year was administered. Adolescents over age 18 were seen in adult primary care settings. Since pediatric and adult clinical workflows may differ, these results are presented separately. Procedure
Author Manuscript
For each calendar year (2010, 2011, 2012) the HMO created two series of patient data summaries. The first series focused on patterns of PHQ-9 use and depression diagnoses associated with PHQ-9 results. The steps were as follows: identify all unique adolescents who were administered the PHQ-9; within the group of unique adolescents with a PHQ-9 result, identify those whose PHQ-9 score was above the clinical cutoff; within this group, identify all adolescents with an incident positive PHQ-9 result; within this group, identify how many adolescents have subsequent new diagnoses of depression. The second series provided context for the first by comparing the number of depression diagnoses associated with PHQ-9 results against the total number of depression diagnoses made at the HMO in the selected departments. The steps were: identify all unique adolescents with a diagnosis of depression in each calendar year; within this group, identify all unique adolescents with a depression diagnosis that had an associated incident positive PHQ-9. Chi-square tests assessed the significance of changing patterns of PHQ-9 use, frequency and departmental location of positive PHQ-9 results, apparent incident depression symptoms, and new diagnosis.
Author Manuscript
RESULTS Use of PHQ-9 in primary care The number of PHQ-9 administrations by department changed significantly from 2010 to 2012 (χ2=1514.74, df=2, p
Psychiatr Serv. Author manuscript; available in PMC 2017 August 15. Published in final edited form as: Psychiatr Serv. 2016 June 01; 67(6): 636–641. doi:10.1176/appi.ps.201400465.
Screening and diagnosis of depression in adolescents in a large HMO R. Eric Lewandowski, Ph.D., Department of Child and Adolescent Psychiatry, New York University, New York
Author Manuscript
Briannon O’Connor, Ph.D., Department of Child and Adolescent Psychiatry, New York University, New York Andrew Bertagnolli, Ph.D., Care Management Institute, Kaiser Permanente, Oakland, California Aldo Tinoco, M.D., National Committee for Quality Assurance, Washington, D.C Arne Beck, Ph.D., Institute for Health Research, Kaiser Permanente of Colorado, Denver William Gardner, Ph.D., Research Institute of Children’s Hospital of Eastern Ontario, Ottawa, Ontario, Canada Department of Pediatrics, Ohio State University, Columbus
Author Manuscript
Christine Jelinek-Berents, M.D., Department of Pediatrics, Kaiser Permanente of Colorado, Denver Doug Newton, M.D., Care Management Institute and the, Department of Child and Adolescent Psychiatry, Kaiser Permanente of Colorado, Denver Kris Wain, M.S., Institute for Health Research, Kaiser Permanente of Colorado, Denver Jennifer Boggs, M.S., Institute for Health Research, Kaiser Permanente of Colorado, Denver
Author Manuscript
Nancy Brace, R.N., Institute for Health Research, Kaiser Permanente of Colorado, Denver Patricia De Sa, Ph.D., Care Management Institute, Kaiser Permanente, Oakland, California Sarah H. Scholle, Dr.P.H., National Committee for Quality Assurance, Washington, D.C Kimberly Hoagwood, Ph.D., and
Disclosures The authors report no conflicts of interest or other disclosures.
Lewandowski et al.
Page 2
Department of Child and Adolescent Psychiatry, New York University, New York
Author Manuscript
Sarah M. Horwitz, Ph.D. Department of Child and Adolescent Psychiatry, New York University, New York
Abstract OBJECTIVE—The aim of this analysis was to determine changing patterns of depression screening and diagnosis over three years in primary and specialty mental health care in a large health maintenance organization (HMO) as part of a quality measure development project.
Author Manuscript
METHODS—Two series of aggregate data spanning three years (2010–12) were gathered from the electronic health record of the HMO summarizing unique adolescents age 12–21 (N=44,342) who had visits in primary and mental health care. Chi-square tests assessed the significance of changing frequency and departmental location of PHQ-9 administration, incident depression symptoms, and depression diagnoses. RESULTS—There was a significant increase in PHQ-9 use from 2010 to 2012, predominantly in primary care, consistent with internally generated organizational recommendations to increase screening with the PHQ-9. The increase in PHQ-9 use led to an increase in depression diagnoses in primary care and shift in the location of some diagnoses from specialty mental health care to primary care. The increase in PHQ-9 use was also linked to an increase in the proportion of positive PHQ-9 results not leading to formal depression diagnoses.
Author Manuscript
CONCLUSIONS—The rate of depression screening in primary care increased over the study period. This increase corresponded to an increase in the number of depression diagnoses made in primary care and a shift in the location of depression diagnoses made from the mental health department to primary care. The frequency of positive PHQ-9 administrations not associated with depression diagnoses also increased. Depression affects 12 to 25% of adolescents,1,2 and is associated with a range of negative academic, social, and health outcomes.3–6 Despite the high burden of depression and the availability of effective treatments, approximately 60% to 80% of affected adolescents do not receive appropriate care.7–9 Pediatric primary care is an important site for the identification of depression in adolescents, the critical first step in connecting youth to treatment.10,11 However, depression remains poorly identified in this setting.12–15 Professional organizations and expert consensus in pediatrics and psychiatry have recommended routine depression screening of adolescents to improve case identification, though available evidence suggests that screening coverage is very low.16–18
Author Manuscript
Quality measurement is an increasingly prominent approach for improving care, whereby providers and organizations are incentivized to adhere to essential healthcare practices.19–22 Depression screening in primary care has been recommended by the United States Preventive Services Task Force (USPSTF) and other major practice guidelines, and has featured prominently in efforts to improve depression care,12,15 including in national initiatives to develop quality measures. For example, the Center for Medicare and Medicaid Services (CMS) included a quality measure of depression screening and follow up for adolescents and adults in the Electronic Health Record Incentive program for “meaningful
Psychiatr Serv. Author manuscript; available in PMC 2017 August 15.
Lewandowski et al.
Page 3
Author Manuscript
use” of electronic health records (EHR). Similarly, the National Collaborative for Innovation in Quality (NCINQ), one of seven centers of excellence funded under the Agency for Health Research and Quality and CMS Pediatric Quality Measures Program developed a suite of potential quality measures organized around a care pathway for managing adolescent depression in primary care in which the first step, and first potential quality measure, is screening for depression.22
Author Manuscript
METHOD
Author Manuscript
As the next step in quality measure development, NCINQ partnered with a large HMO to study changing patterns of depression screening over time. This study also assessed depression diagnoses following screening, which corresponds to the second measure in the NCINQ care pathway, and directly addresses the specific research need highlighted by USPSTF to link screening to improved identification of depression cases.16 Large scale, naturalistic data on screening and diagnosis will help to identify gaps in essential care practices related to depression case identification, and provide an indication of the fit and possible value of the depression quality measures in the context of current practice.
Measures
Setting and data sources Data for adolescents meeting inclusion criteria were abstracted from the EHR of a large HMO. The identity of the HMO has been masked per agreement with the organization. The Chesapeake Institutional Review Board determined that this study was not human subjects research since only de-identified, aggregated data were collected. Starting in 2011, the HMO began to implement the adolescent version of the Patient Health Questionnaire-9 (PHQ-9) as a screening tool in primary care.23, 24 Previously, the PHQ-9 was used primarily in the mental health department for diagnostic support and monitoring of patients with known diagnosis of depression. Computer programmed extraction of aggregate data for this study was conducted from May 29–31, 2013.
Author Manuscript
The Patient Health Questionnaire-9-Modified form—The PHQ-9 is a 9-item selfreport questionnaire assessing depression symptoms and severity that has been validated its original form with adolescents.23,24 Items based on the DSM-IV criteria for depression are rated on a 4-point scale ranging from 0 (“not at all”) to 3 (“nearly every day”). The PHQ-9Modified includes minimal adjustments to the original PHQ-9 to incorporate characteristics of depression in adolescents and age-appropriate language. Specifically, the PHQ-9Modified includes irritability in the item assessing depressed mood, and weight loss to the item assessing appetite. No psychometric data is available for the PHQ-9-Modified, but because it is identical to the PHQ-9, besides the described adjustments, the developers indicate that using PHQ-9 cutoffs is appropriate. Throughout this paper “PHQ-9” is used to indicate the PHQ-9-Modified. Consistent with research literature and practices within the HMO, a PHQ-9 score of 11 or greater indicated a positive screen.24 In this study, a positive PHQ-9 score that was preceded by a period of 6 months with no record of depression diagnosis or antidepressant medication ordered, was termed an “incident positive PHQ-9” (distinguishing it from repeated PHQ-9 administrations with the same patient to track
Psychiatr Serv. Author manuscript; available in PMC 2017 August 15.
Lewandowski et al.
Page 4
Author Manuscript
symptoms in a prevalent case) and taken to represent the identification of new depression symptoms. The 6-month depression-free period preceding a positive PHQ-9 began 14 days prior to the PHQ-9 chart entry date to allow for institutional lag time between PHQ-9 administration and upload into the EHR. Depression diagnosis—Depression diagnoses in the current study included ICD-9 codes for Major Depressive Affective Disorder, Depressive-type Psychosis, Depressive DisorderNOS, Adjustment Disorder with Depressed Mood, with Mixed Anxiety and Depressed Mood, with Mixed Disturbance of Emotions and Conduct.25 A new depression diagnosis was defined as one preceded by a period of 6 months with no depression diagnosis or antidepressant medication order within the EHR. Study Group
Author Manuscript
For three years (2010, 2011, and 2012), adolescents were included in the study sample if they were 12 to 21 years of age on December 31st of the previous year and if their EHR contained documentation of at least one face-to-face visit with a provider. Adolescents may have had multiple visits and multiple PHQ-9 results in a given calendar year, but were counted only once. Adolescents receiving more than one PHQ-9 in a given calendar year were counted in the department where the first PHQ-9 of the year was administered. Adolescents over age 18 were seen in adult primary care settings. Since pediatric and adult clinical workflows may differ, these results are presented separately. Procedure
Author Manuscript
For each calendar year (2010, 2011, 2012) the HMO created two series of patient data summaries. The first series focused on patterns of PHQ-9 use and depression diagnoses associated with PHQ-9 results. The steps were as follows: identify all unique adolescents who were administered the PHQ-9; within the group of unique adolescents with a PHQ-9 result, identify those whose PHQ-9 score was above the clinical cutoff; within this group, identify all adolescents with an incident positive PHQ-9 result; within this group, identify how many adolescents have subsequent new diagnoses of depression. The second series provided context for the first by comparing the number of depression diagnoses associated with PHQ-9 results against the total number of depression diagnoses made at the HMO in the selected departments. The steps were: identify all unique adolescents with a diagnosis of depression in each calendar year; within this group, identify all unique adolescents with a depression diagnosis that had an associated incident positive PHQ-9. Chi-square tests assessed the significance of changing patterns of PHQ-9 use, frequency and departmental location of positive PHQ-9 results, apparent incident depression symptoms, and new diagnosis.
Author Manuscript
RESULTS Use of PHQ-9 in primary care The number of PHQ-9 administrations by department changed significantly from 2010 to 2012 (χ2=1514.74, df=2, p
