The Applicability And Equivalence Of The Patient Health Questionnaire And Hospital Anxiety And Depression Scale In Patients With Major Depression

BackgroundMajor Depression is globally prevalent.In clinical practice,it takes challenges in detection and diagnosis because the symptoms of depression can present different forms of combinations.Well-qualified rating scales for mental health could be used in measuring depression and concurrent anxiety symptoms for the purpose of guiding decision making in treatment.Given that different instruments produce scores on different metrics,it is hard to compare results of different instruments directly.So the clinicians and investigators need a crosswalk to transform raw scores from one instrument to the other across situations that use either one.The Rasch model is derived from the Item Response Theory(IRT),which could be utilized to equating instruments and generating their crosswalk.The Patient Health Questionnaire(PHQ)and the Hospital Anxiety and Depression Scale(HADS)are widely used self-rating scale to screen depression and anxiety symptoms.The PHQ contains a depression scale—9-item Patient Health Questionnaire(PHQ-9)and an anxiety scale—7-item Generalized Anxiety Disorder Scale(GAD-7).The HADS are divided into anxiety subscale(HADS-A)and depression subscales(HADS-D).Analyzing and comparing there scales,generating their crosswalks might improve efficiency in clinical assessment facilities.Objective(1)Applying the Rasch model to analyze and compare the psychometric properties of PHQ and HADS in order to investigate their applicability in patients with major depression.(2)Applying the Rasch model to equate the PHQ-9 to HADS-D as well as the GAD-7 to HADS-A,establishing their crosswalks and verify the effectiveness.MethodsA total of 460 patients who aged from 18 to 63 years old and met the DSM-5 diagnostic criteria of major depressive disorder were recruited at the psychiatric department of a general hospital.The patients were interviewed to collect their demographic characteristics,then finished the PHQ and HADS on their own.Using Rasch analysis to examine the dimensionality,fit,reliability,category structures and Differential Item functioning(DIF)of the scales,comparing their Item-person map and Test information curve(TIC).A set of preliminary analyses were adopted to ensure the data to be appropriate for the linkage.Afterwards,adopting common-person equating method to establishing the crosswalks of PHQ-9/HADS-D as well as GAD-7/HADS-A.Conducting paired-samples T-test and Intraclass correlation coefficient(ICC)to validate the crosswalks.ResultThe principal components analysis verified that the HADS had a two-dimensional structure,which contained anxiety and depression dimension.All the PHQ-9,GAD-7,HADS-D and HADS-A were proved to be with unidimensional structure(eigenvalue for first contrast<2.0)and good reliability(reliability index>0.8).All their items fit the model(Infit MnSq/Outfit MnSq ranged from 0.5 to 1.5).No disordering category was found.All the item difficulties were concentrating on middle level and well-targeted by person abilities that no significant ceiling or floor effect was found.Item PHQ9 and GAD6 were found to be with gender-related non-uniform DIF,female on high ability levels had advantages.The means of PHQ-9 and HADS-D were-0.11±0.58logits and 0.15±0.26logits respectively,PHQ-9 was easier than HADS-D.The TIC of PHQ-9 and HADS-D exhibited the same wide range of symptoms severity,but more information the PHQ-9 contributed.The means of GAD-7 and HADS-A were-0.14±0.54logits and 0.14±0.63logits respectively,GAD-7 was easier than HADS-A.The test information provided by GAD-7 is superior to the HADS-A across a wider range on low levels of anxiety severity,while they contributed much the same information.In the PHQ-9/HADS-D crosswalk,the differences between raw scores and converted scores were not significant(p>0.05),nor in the GAD-7/HADS-A crosswalk(p>0.05).The ICC between raw scores and converted scores were all in the range of 0.86～0.91,exhibiting good consistency.Conclusion(1)Psychological properties of the PHQ and HADS were proved to be acceptable to apply in patients with major depression.Both of them exhibited better discriminability and precision when assessing individuals at middle levels of symptom severity than at high or low levels.The PHQ-9 performed higher precision than HADS-D and the GAD-7 effectively assessed a wider range on low levels of anxiety severity than HADS-A.Overall,the PHQ was superior to the HADS.(2)Two items exhibited bias of gender.At high levels of symptoms severity,females were slightly more likely than males to endorse item PHQ9(suicidal ideation)and GAD6(Annoyance or irritability).(3)The crosswalks of PHQ-9/HADS-D as well as GAD-7/HADS-A were successfully established and proved to be validity.