Table 2.
Application of Wilson and Cleary model
| Characteristics of Study | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Author Year Country | Population | Design | Latent factors/measure | Sample size | Age Mean (SD) | % of Female | Aim of study | Analytical Tool | Results/Findings | Percentage of variance explained by model |
| Ade-Oshifogun 2012 USA | Obesity/Chronic Pulmonary Disease (COPD) | Cross sectional | BP: BMI, FEV1, DLCO, Percent trunk fat (DEXA) SS: Dyspnoea (CRQ), fatigue (CRQ), sleep apnoea (ESS) FS: 6-min walk distance (6MWD) GHP: Functional Performance Inventory (FPI) |
76 | 69.7 (10.3) |
35.5% | To test a theoretically and empirically supported model of the relationship among clinical variables, symptoms, function status and health status of elderly people with COPD | Path analysis | ● Function status, symptoms and biological variable DLCO have direct causal effect on health status ● DLCO ad dyspnoea predict functioning ● The effect of clinical variables on health status is mediated by symptoms ● Symptoms, function status and clinical variable indirectly influence health status |
● Model explained 29% of the variance ● Clinical variables explain 29.6% of symptoms ● Clinical variables explained 50.5% of function status |
| Arnold 2005 Netherlands |
1. Chronic Obstructive Pulmonary Disease (COPD) 2. Chronic Heart Failure (CHF) |
Cross sectional | BP: COPD: FEV1
VHF: LVEF SS: Dyspnoea measured by a questionnaire FS: Physical Functioning subscale of SF-36 GHP: General health subscale of SF-36 HRQL: Perceived health competence scale |
COPD:95 CHF 90 |
65 (9.3) 59 (10) |
35.8% 24.4% |
To investigate relationship between objective and subjective health in patients with COPD and CHF | Structural equation model (SEM) | ● Biological/physiological variables in both diseases are not significantly related to symptoms but predict physical functioning for COPD (β = 0.20) and CHF (β = 0.17) ● Symptoms predict physical functioning in COPD (β = 0.63) and in CHF (β = 0.67). ● Physical functioning associate with general health perceptions in COPD (β = 0.39) and CHF 9 β = 0.32) ● Symptoms directly associate with general health perceptions only in COPD ● In COPD, symptoms, physical functioning explain general health perception ● Only physical functioning explains general health perceptions in CHF |
● Global HRQL explained by symptoms and general health perceptions in both diseases. |
| Baker 2007 UK |
Xerostomia | Longitudinal | BP: Salivary flow Clinical signs SS: Xerostomia Inventory (XI) FS: (OHIP-14) GHP: Global oral health rating (GOH) HRQL: (HADS) |
85 | 59.8 (11.5) | 76.5% | To systematically test Wilson and Cleary conceptual model of the direct and mediated pathways between clinical and non-clinical variables in relation to the oral health-related quality of life (OHRQoL) of patients with xerostomia. | Structural Equation Modelling (SEM) | ● More severe clinical signs were associated with worse patient-reported symptoms ● More symptoms predicted a greater impact on everyday oral functioning ● Worse functioning predicted lower global oral health perceptions ● Both biological indicators and functioning predicted subjective well-being |
● Function accounted for 96.9% of total effects ● 88.2% of total effect on functioning was mediated by symptoms status ● Symptoms 9% ● Functioning 22% ● GOH 24% ● Well-being 21% |
| Brunault 2014 France |
Obesity | Cohort | BP: BMI Type of Surgery SS: BDI Bulimic Investigatory Test, Edinburg (BITE) FS: Quality of Life, Obesity and Dietetics (QOLOD) -Physical QoL -Psychological QoL -Social QoL -Sexual QoL -Comfort with food |
126 | 40.2 (10) | 79.4% | To put the Wilson Cleary model to test by determining the predictors of postoperative change in each QoL dimension 12 months after bariatric surgery | Linear mixed model | ● Improvement in Psychosocial QoL was associated with lower preoperative depression severity, lower preoperative binge eating severity and higher weight loss ● Improvement in Sexual QoL was associated with lower preoperative depression severity, lower preoperative binge eating severity and younger age ● Improved comfort with food was associated with lower preoperative binge eating severity |
● ? |
| Carlson 2014 USA |
Heart Failure | Cross-sectional | BP: Number of chronic illness Comorbidity burden (CCI)as in index of severity of illness Diagnosis of diabetes Diagnosis of chronic atrial fibrillation SS: Depression measure with PHQ-9 Physical symptoms measured with KCCQ FS: Physical and social functioning measured with KCCQ GHP: First item in the SF-36(v2) |
265 | 62 | 35.8% | To determine the key predictors of overall perceived health (OPH) | Hierarchical multiple regression | ● Age, gender and race/ethnicity were predictors of OPH ● Perceived sufficiency of income, social functioning, comorbid burden, symptom stability, black compared to white race were independent predictors of OPH ● Physical and social functioning mediated the effect of SOB and fatigue on OPH as well as the effect of symptom on OPH |
● 39.2% |
| Cosby 2000 USA |
HIV/AIDS | BP: CD4 counts SS: Health distress, mental health, energy/fatigue and pain of Health Status Questionnaire (HSQ), SSC-HIV FS: Physical, role, social and cognitive functioning of HSQ GHP: QAM, General health perception of HSQ HRQL: Overall quality of life of HSQ |
146 | To determine the relationships among haematological complications associated with AIDS, characteristics of the individual and the five dimensions of Wilson and Cleary model | Logistic regression | ● All five dimensions of Wilson and Cleary model significantly predicted anaemia. | ||||
| Eilayyan 2015 Canada |
Asthma | Longitudinal | SS: Physical symptoms (MAQLQ-symptoms) Emotional symptoms (MAQLQ-emotion) Self-efficacy (KASE-AQ) FS: Physical function (MAQLQ-activity) |
299 | 62.1 (14.4) | 69% | To identify direct and indirect predictors of perceived asthma control among primary care population. | Path model | ● Symptom was affected by self-efficacy ● Emotional status was affected by symptom and self-efficacy ● Physical activity was affected through symptom, emotional status and self-efficacy ● Perceived asthma control at baseline was affected by asthma symptom, physical activity, self-efficacy and smoking ● Perceived asthma control at follow-up was predicted by asthma symptom, physical activity, self-efficacy and baseline perceived asthma control. ● Perceived asthma control was indirectly predicted by emotion status through self-efficacy and physical activity |
|
| Halvorsrud 2010 Norway |
Chronic Disease | Cross- sectional | SS: Geriatric Depression Score (GDS-15) FS: SF-12 subscale of physical function GHP: Health satisfaction: global item measure from WHOQoL-Bref HRQL: WHOQoL-Old |
89 | 78.6 | 73% | To explore the predictors of QOL among community-dwelling older adults receiving community health care | Path analysis: Structural equation Modelling (SEM) | ● Environment has direct effects on QOL and indirect effects on QOL with depressive symptoms and health satisfaction (GHP) as mediators ● Depressive symptoms had an indirect, negative effects on QOL with physical functions and general health perceptions as mediators ● Health satisfaction was a mediator between physical function and QOL |
● The predictor variables accounted for 37% of the variance in depressive symptoms, 29% in physical function, 44% in general health perceptions and 66% of the variance in QOL (the overall model) |
| Heo 2005 USA |
Heart failure | Baseline data | BP: Patient interview Medical records, CCI SS: Patients perception of Presence and severity of dyspnoea and fatigue measured by Dyspnoea-Fatigue Index Questionnaire FS: NYHA GHP: SF-36 HRQL: MLHFQ |
293 | 73 (11) | 53% | To determine the bivariate relationships between HRQL and other variables proposed by Wilson and Cleary To determine best multivariate model based on these variables To test specific components of the Wilson and Cleary model of HRQL |
Multiple regression | ● Health perception, symptom status and age predict HRQL ● Health perception mediates the effect of symptoms on HRQL ● Functional status does not mediate the effect of symptom status on health perception |
● Final model explains 29% of the variance |
| Hofer 2005 Austria |
Coronary Artery Disease (CAD) | Longitudinal | BP: Severity of CAD (no of diseased vessel No. of risk factors SS: Canadian Cardiovascular Society classification of angina pectoris FS: SF-36 physical function score GHP: SF-36 general health score HRQL: Scores on the three scales (physical, social and emotional) of MacNew Heart Disease Quality of Life Questionnaire |
432 | 61.8 (10.2) | 24.1% | To apply Wilson and Cleary model a priori to patients with CAD in a prospective longitudinal design and to find out whether it is applicable to CAD patients and is stable over time. | Structural Equation Modelling (SEM) | ● Physical functioning, anxiety symptoms have effect on overall HRQL ● Anxiety predicts poorer HRQL ● Depression affects physical functioning and general health perception. ● The higher the level of anxiety, the more severe the symptoms reported |
● Final model explains 49% at baseline, 62% one month after and 66% 3 months after intervention of the variance of overall HRQL |
| Kanters 2012 Netherlands |
Pompe disease | Cross-sectional | BP: Enzyme activity (fibroblasts) Skeletal muscle strength assessed by MRC, respiratory function assessed by FVC SS: shortness of breath, Fatigue assessed by Fatigue Severity Scale (FSS) FS: Rotterdam Handicap Scale (RHS) GHP: EQ-5D Visual Analogue Scale (EQ-5D-VAS) HRQL: MCS and PCS of SF-36 Utility derived from EQ-5D |
103 | 49.3 | 50.6% | To develop a conceptual model for Pompe disease in adults and statistically test it in untreated patients | Random effects linear regression | ● MRC and FSS were negatively associated with disease duration ● FVC was affected by female gender ● RHS was affected by FSS, MRC, FVC and Age ● EQ-5D Vas was associated with RHS and disease duration ● MCS was associated with EQ-5D VAS ● PCS was associated with EQ-5D VAS ● Utility was associated with EQ-5D Vas |
|
| Krethong 2008 Thailand |
Heart Failure | Cross- sectional | BP: Medical records-LVEF SS: Cardiac Symptoms Survey (CSS) FS: NYHA functional classification GHP: 100 mm horizontal visual analogue scale HRQL: MLFHQ |
422 | 58.47 | Ns | To develop and test a hypothesized causa model of HRQL in Thai heart-failure patients | Structural equation modelling (SEM) | ● Biological/physiological affected functional status (β = −0.34, p < 0.05). ● Symptom affected functional status (β = 0.45, p < 0.05); GHP (β = −0.27, p < 0.05) and HRQL (β = −0.48, p < 0.05) ● Functional status had impact on GHP (β = −0.28, p < 0.05); HRQL (β = −0.25, p < 0.05) ● Social support had impact on symptom (β = −0.25, p < 0.05); GHP (β = 0.19, p < 0.05) and HRQL (β = −0.17, p < 0.05) ● The effect of biological/physiological on symptom was not significant. |
Model explained 58% of the variance in overall HRQL |
| Mathisen 2007 Norway |
Heart Surgery | Longitudinal | GHP: General Health subscale of SF-36 HRQL: Global Quality of Life (gQoL) Norwegian version of the Quality of Life Survey (QoLS-N) |
108 | 64.2 | 19% | To investigate the existence of a reciprocal relationship between patients’ assessment of quality of life and their appraisal of health. | Structural equation modelling (SEM) | ● Baseline overall QoL has a cross lagged effect on three months assessment of general health ● The path from general health at six months to QoL at 12 months was significant ● The simultaneous effects model demonstrated a bidirectional causal paths at each point observed after baseline |
|
| Mayo 2015 Canada |
Stroke | Cross-sectional | BP: Side of lesion Stroke severity measured with CNS, CCI SS: SIS Pain: SF-36 (body pain) Vitality: SF-36 (vitality) Emotional well-being: SF-36 (mental health) FS: Physical Functioning: SF-36 (PF) SIS (mobility) Health Utility Inventory(HUI): HUI (ambulation) HUI (dexterity) Social Functioning: SF-36 (SF) SIS 8b Role: Worst of SF-36 RE & RP Cognitive: Mini mental State Education (MMSE) GHP: EQ-5D VAS SF-36 (General health) |
678 | 67.3 (14.8) | 45% | To empirically test a biopsychosocial conceptual model of HRQL for people recovering from stroke | Structural equation modelling (SEM) | ● Less comorbidity, less pain, better memory and more vitality associated with better health perception. | |
| Nokes 2011 USA |
HIV/AIDS | Cross sectional | SS: Centre for Epidemiological Depression Scaled (CES-D) Revised SSC-HIV Body Change Distress Scale HRQL: HAT-QOL |
1217 | 41.7 (9.1) | 31.5% | To determine if there were age-related differences in symptoms status and HRQL for HIV-positive persons aged 50 years and older compared with younger (aged 49 years and younger). | Stepwise regression | ● Age was a predictor for sexual function and provider trust ● Less depressive symptoms and less body change distress were related to increase in sexual functioning |
|
| Phaladze 2005 Sub-Saharan Africa | HIV/AIDS | Cross sectional | BP: Has been given AIDS diagnosis Has Comorbidities SS: Revised SSC-HIV FS: Overall functioning GHP: Health worries HRQL: HAT-QOL. |
743 | 34.1 (9.6) | 61.2% | To increase understanding of the meaning of quality of life for people living with HIV/AIDS in four countries in Sub-Saharan Africa: Botswana, Lesotho, South Africa and Swaziland. | Hierarchical multiple regression | ● Daily functioning predicts overall HRQL ● Higher level of education associates with lower HRQL ● Higher symptom intensity associates with lower HRQL ● A close correlation between symptom intensity and functional status |
● Overall model explains 53.2% of the variance |
| Portillo 2005 USA |
HIV/AIDS | Cross sectional | BP: Has been given AIDS diagnosis Has Comorbidities SS: Revised SSC-HIV FS: Overall functioning GHP: Health worries (HAT-QOL) |
920 | 41 (8.7) | 32.6% | To test the Wilson and Cleary model in a sample of ethnic minority persons living with HIV/AIDS | Hierarchical regression | Association between physiologic factors, symptoms, functioning, general health perception and life satisfaction | ● Overall model explains 22.9% |
| Saengsiri 2014 Thailand |
Coronary Artery Disease (CAD) | BP: LVEF Rose Questionnaire for angina Rose Dyspnea Scale (RDS) SS: Centre for Epidemiologic Studies Depression Scale (CES-D) Cardiac Self Efficacy Scale (C-SES) FS: Functional Performance Inventory Short-Form (FPI-SF) SF-36 Vitality subscale HRQL: Quality of Life Index-Cardiac Version |
303 | 61.2 (10.9) | 26.4% | To explain relationship between cardiac self-efficacy, social support, biological and physiological (LVEF) symptoms of angina, dyspnoea, depression, vital exhaustion, functional performance and quality of life in post-PCI CAD patients | Pearson Correlation Path analysis | ● Social support (β = 0.31), depression(β = 0.24), vital exhaustion (β = 0.23) and cardiac self-efficacy(β = 0.21) had the most powerful direct effect on quality of life of post-PCI CAD patients ● Self-efficacy had indirect effect on quality of life (β = 0.21, p < 0.001) |
||
| Santos 2015 Brazil |
Oral health | Cross sectional | BP: Edentulism (dentate = 0, edentulous = 1) assessed by clinical examination SS: Assessed using the question, “are you satisfied with the appearance of your prostheses?” FS: Assessed with the question, “have you decreased or changed the type of food because of problems with your teeth or dental prostheses?” GHP: Assessed using the question, “compared with others your age, how would you rate the health of your mouth overall?” HRQL: OHIP-14 |
578 | 68 (6.3) | 67.3% | To test the Wilson and Cleary model of the direct and mediated pathways between clinical and non-clinical variables in relation to oral health-related quality of life | Structural Equation Modelling (SEM) | ● Dissatisfaction with symptom status are associated with worse functional status ● Worse functioning predicts poor oral health perception ● Poor oral health perception associates with higher worse oral health quality of life ● Final model shows negative significant direct effect between biological variable and symptom status ● Age, gender and geographical location have direct paths to biological variable (edentulism) ● Age and gender directly impact oral health-related quality of life |
● The comparative fit index is 0.98 indicating adequate fit. |
| Schulz 2012 Netherlands |
Kidney Transplant | Cross-sectional | BP: Number of active comorbidities reported by patients FS: European Quality of Life −5 dimension (EQ-5D) GHP: EQ-5D Visual Analogue Scale (EQ-5D-VAS) HRQL: General Health Questionnaire (GHQ-12) |
609 | 53.7 (12.3) | 43.9% | To identify pathways through which objective health affects psychological distress and to clarify how personal characteristics are shaped by objective health and determine psychological distress | Structural equation modelling (SEM) | ● Impact of objective health and functional status on psychological distress was fully mediated by subjective health and personal characteristics ● Influence of objective health was mediated by successively by functional status and personal characteristics; successively by functional status and subjective health; exclusively by personal characteristics and; exclusively by subjective health |
The model explained 32% of variance of psychological distress |
| Shiu 2014 Hong Kong |
Diabetes | Cross sectional | BP: Time since diagnosis Age of onset and type of diabetes HbA1c level, blood pressure and lipid profile SS: Self-reported comorbidity characteristics and presence of comorbidity and no of comorbidities FS: Physical functioning subscale of SF-36 Older American Resources and Services Multidimensional Functional Assessment Questionnaire GHP: SF-36: general health Self-developed ratings 6 HRQL: subscales of the SF-36: role-physical, role-emotional, mental health, social functioning, bodily pain and vitality |
452 | 71.8 (7.3) | 59.1% | To apply the Wilson and Cleary model of HRQL to understand the relationship among clinical and psychological outcomes in community-dwelling older Hong Kong Chinese people with diabetes. | Structural Equation Modelling (SEM) | ● Four determinants: general health perception, psychological distress, adequacy of income and social support have direct effect on HRQL ● Three determinants: symptom status, physical functional status and psychological status have indirect effects on HRQL through general health perception ● Four determinants: symptom status, age, gender and physical activity have indirect effect on HRQL through physical function status |
● The model explains between 64% and 72% of variance |
| Sousa 1999 USA |
HIV/ AIDS |
Cross- sectional | BP: APACHE III SS: HIV-problem checklist FS: HIV Quality Audit marker (QAM) GHP: MOS-30 (single item for GHP) HRQL: MOS-30 (single item for overall quality of life |
142 | 38 (8.7) | 20% | Multiple regression | ● Symptoms correlated negatively with GHP (r = −0.48) and overall HRQL (r = −0.37). Functional status positively associated with GHP (r = 0.22) and overall HRQL (r = 0.29) Biological/physiological variables do not have significant associations either directly or indirectly on any of the variables. | ● | |
| Sousa 2006 USA |
HIV/ AIDS |
Cross- sectional | BP: CD4 Count SS: SSC-HIV FS: The Health Assessment Questionnaire-Disability Index (HAQ-DI) GHP: 100 mm visual analogue scale Ordinal scale HRQL: Derived from general health status scales |
917 | 30.4 (8.13) | 43% | To estimate the primary pathways of the Wilson and Cleary HRQL conceptual model using structural equation modelling (SEM) | Structural equation modelling (SEM) | ● A significant relationship between status and functional health (r = 0.56) ● There is significant relationship between symptoms status and general health perceptions (r = −0.33) and functional health and general health perceptions (r = −0.42) ● There is significant relationship between symptoms status and overall quality of life (r = −0.20) and between GHP and overall quality of life (r = 0.26) CD4 count had a negative relationship with symptom status (r = − 0.20, p < 0.05) |
● Symptoms explain 49% of functional health ● Both symptoms status and functional heath accounted for 62.5% of the variance of general health. ● Both symptoms status and general heath perceptions accounted for 38,2% of the variance in overall quality of life. |
| Ulvik 2008 Norway |
Coronary Artery Disease (CAD) | Cross- sectional | BP: Myocardial disease LVEF SS: Angina (AFS, CCS) Dyspnoea (NYHA) Anxiety (HADS) Depression (HADS) FS: Physical function Social function GHP: General health (SF-36) HRQL: Overall QoL: measured with a single question |
753 | 61.7 (10.2) | 26% | To analyse relationship between disease severity and both mental and physical dimensions of HRQL. | Linear and ordinal logistic regression | ● Biological variables associate with symptoms ● Depression associates positively with LVEF ● Symptoms affect physical function ● Social function is low in patients with more symptoms of anxiety. ● General health is negatively related to anxiety and depression but positively related to physical and social functions ● Better overall QOL is associated with less symptoms and depression but related negatively to social function |
● The model explains 43% of the variance of overall quality of life. |
| Wettergren 2004 Sweden |
Hodgkin’s Lymphoma | Cross sectional | BP: Disease stage (I-IV) Treatment modality (irradiation, chemotherapy or combined modality treatment Time since diagnosis SS: (SEQoL-DW) HADS FS: Measured as part of general health perceptions GHP: PCS of Short Form 12 (SF-12), MCS of SF-12 HRQL: QoL index of (SEQoL-DW) |
121 | 45 (median) | 45% | To evaluate HRQL in long-term survivors of |Hodgkin’s lymphoma (HL) and to identify determinants of HRQL using Wilson and Cleary’s conceptual model with the potential goal of improving care and rehabilitation. | Partial Correlations | ● Disease stage correlated with Disease index (SEQoL-DW) ● Lower SOC was related to a worse HRQL ● Poorer physical health was associated with worse overall quality of life. |
|
| Wyrwich 2011 USA | General Anxiety Disorder (GAD) | Longitudinal | BP: CGI-S SS: HAM-A FS: PSQI GHP: Q-LES-Q(SF) (items 1–14) HRQL: Q-LES-Q(SF)) (Item 16) |
1692 | 40.3 (11.8) | 65.1% | To test the application of the Wilson-Cleary model to patient population with generalised anxiety disorder (GAD) using longitudinal clinical trial data. | Path Model | ● CGI-S had a strong relationship with HAM-A ● HAM-A at week 8 had strong path (β = 0.5) to PSQI and moderate effect (β = −0.40) on Q-LES-Q(SF) ● Q-LES-Q(SF) had a strong relationship with overall quality of life (β = 0.66) |
● Model explained 56% at baseline and 69% at week 8 |
DLCO Carbon Monoxide Diffusing Capacity, FEV1 Forced Ejection Volume, FVC Forced Vital Capacity, PSQI Pittsburgh Sleep Quality Index, LVEF Left Ventricular Ejection Fraction, QAM Quality Audit Marker, CCI Charlson Comorbidity Index, OHIP-14 Oral Health Impact Profile, KCCQ Kansas City Cardiomyopathy Questionnaire, MCS Mental Component Summary, BDI Beck Depression Index, PHQ-9 Patient Health Questionnaire, HAM-A Hamilton Rating Scale for Anxiety, MRC Medical Research Council, CNS Canadian Neurological Scale, SIS Stroke Impact Scale HAT-QOL, HADS Hospital Anxiety and Depression Scale, BMI Body Mass Index, PCS Physical Component Summary, HSQ: Health Status Questionnaire, CRQ Chronic Respiratory Disease Questionnaire, MLFHQ Minnesota Living with Heart Failure Questionnaire, NYHA New York Heart Association, SEQoL-DW Schedule for the Evaluation of the Individual Quality of Life Direct Weighting, CGI-S Clinical Global Impression-Severity of Illness, Q-LES-Q(SF) Quality of Life, Enjoyment and Satisfaction Questionnaire-Short Form, HIV/AIDS Targets Quality of Life, SSC-HIV-Signs and Symptoms Checklist for Persons with HIV/Disease, WHOQOL World Health Organisation Quality of Life