Abstract
Background
Self-management is an important measure to control the development of chronic obstructive pulmonary disease (COPD), but the self-management ability of newly diagnosed COPD patients can not be evaluated. Therefore, this study aims to develop and verify a risk prediction model based on the information-motivation-behavioral skills (IMB) model to predict poor self-management behaviors in newly diagnosed COPD patients.
Methods
In this prospective cohort study, a total of 331 adults with COPD were recruited from a general hospital in Chengdu, China. Data were collected at baseline based on the IMB model, such as cognitive function, social support, frailty, depressive and anxiety symptoms, and patient activation. Self-management behaviors were evaluated as the outcome variable after one-year follow up. Multivariate logistic regression was used to develop a risk prediction model to predict poor self-management behaviors. The nomogram was used to perform and visualise the predictive model and the receiver operator characteristic (ROC) curve, external validation were applied to evaluate the prediction performance of the model.
Results
A total of 331 patients completed follow-up (222 in the development cohort and 109 in the validation cohort). 68.3% of the participants occurred poor self-management behaviors. Cognitive function, patient activation, and depression were independent predictors for poor self-management behaviors for COPD patients. A nomogram was established based on regression analysis, and the AUC of this nomogram was 0.945. The sensitivity and specificity were 89.68% and 91.04% respectively. The AUC of the validation cohort was 0.898 and the Hosmer-Lemeshow test indicated good model prediction.
Conclusions
The risk prediction model based on IMB model and a nomogram including 3 easily available prediction factors (cognitive function, patient activation and depression) on poor self-management behaviors for newly diagnosed COPD patients was established, which showed good discrimination, and calibration. It can be used to screen out high- risk population with poor self-management behaviors for newly diagnosed COPD patients early.
Keywords: Risk prediction model, Self-management, Newly diagnosed, COPD
Background
Chronic Obstructive Pulmonary Disease (COPD) is a chronic respiratory disease characterized by limited airflow and incomplete reversibility, with high morbidity and mortality, which seriously threatens human life and health [1]. In 2018, The results of a large-scale cross-sectional study covering 24% of the population in China showed that the prevalence of COPD was 13.6%, which increased by 5.4% compared to 2004 among the included 66,752 adults aged 40 or above [2]. COPD has been identified as an important part of the global disease burden, and projections of national medical costs attributable to COPD are reported to increase to $60.5 billion in 2029 [3]. Therefore, COPD has been widely concerned all over the world.
The Global Initiative for Chronic Obstructive Pulmonary Disease (GOLD) guidelines emphasized the importance of self-management in care of COPD, which could significantly improve the clinical outcomes and quality of life of patients [4]. Self-management behaviours of COPD patients includes the management of medication, oxygen therapy, regular vaccination, timely quitting smoking, healthy diet and regular physical exercise [5]. A structured, personalized and diversified self-management intervention can engage and support the patients to positively adapt their health behaviour(s) and develop skills to better manage their disease [6].
Although the benefits of self-management for COPD patients is well known, the level of self-management ability among COPD patients remains low [7]. In China, 50% of COPD patients were at a low or moderate level of whole self-management ability [8], while the proportion was 86.1% in another study [9]. In the management of medication, a large proportion of COPD patients were found to be failure to follow the drug treatment plan, and nearly half of the patients had poor inhaler technique in O’Conor’s study [10]. In regular physical exercise, another study confirmed that patients with COPD have significantly lower levels of daily physical activity compared to healthy controls [11]. In oxygen therapy, 58.8% had poor adherence to oxygen therapy [12]. The poor self-management behaviours may lead to adverse clinical outcomes, and early identification and intervention for self-management behaviors may effectively improve their quality of life.
However, the current tools for evaluating self-management behaviors are only suitable for COPD patients who have been diagnosed with COPD for some time (not newly diagnosed COPD patients). For example, when using the COPD Self-Management Scale (CSMS), which is widely used to evaluate the self-management ability of COPD patients, it is necessary to ask patients about the frequency of vaccination in one year and how to prevent colds in winter [13]. Therefore, to ensure the accuracy of the assessment results, participants need to be observed for at least one year. However, the newly diagnosed COPD patients are a group that can not be ignored, and increased 49.8% of COPD patients in the world were newly diagnosed from1990 to 2017 [14]. A large cohort study in Chongqing showed that the non-compliance of self-management of newly diagnosed COPD patients was 50% higher than other COPD patients [15]. The possible reason was that some newly diagnosed COPD patients didn’t pay enough attention to self-management due to mild symptoms, or not knowing enough about the disease [15]. For these newly diagnosed patients, poor self-management behaviors may directly increase their readmission rate and mortality, and should be screened out early, so as to carry out intervention as early as possible to minimize the development of disease. But there is no suitable prediction or evaluation tool to evaluate the self-management behaviors of newly diagnosed COPD patients at present.
The information-motivation-behavioural skills (IMB) model is commonly used to enhance understanding of predictive factors for health behaviours and outcomes. The IMB model posits that each element (i.e. information, motivation and behavioral skills) exerts a direct effect on health behaviors [16]. Studies have reported that certain factors such as cognitive function (information); social support, frailty, depressive and anxiety, patient activation (motivation) are associated with self-management behaviors [3, 17]. As we all know, patients with cognitive impairment have difficulties in understanding complex clinical prescriptions and judging the consequences of treatment choices, and may encounter difficulties in dealing with medical information presented to them in rapid oral communication [18]. Therefore, cognitive function is the key to receiving and storing information effectively. In addition, social support, frailty, anxiety and depression, patient activation are the motivation factors for COPD patients to carry out healthy behaviors. Good social support can promote the implementation of healthy behaviors and improve patients’ self-management ability [19]. However, another study confirmed that some patients will weaken their sense of self-management responsibility because of their dependence on social support [20]. Therefore, the impact of social support on self-management was still unclear. Frailty was very common in patients with COPD, and its incidence rate was significantly higher than that of patients with non-COPD of the same age [21]. Frailty was a complicated syndrome, which was characterized by the decline of physiological functions of various organs, and it was difficult for patients with frailty to adhere to healthy behaviors [22]. The prevalence of anxiety and depression in COPD patients was 2 ∼ 4 times that of the general population, and it was difficult for these patients to motivate themselves to participate in executing self-management behaviours [23]. Patient Activation (PA) was first proposed by Hibbard in 2004, which means that patients have the knowledge, skills and confidence to manage their own health [24]. The IMB model, widely applied in chronic disease management, holds that patients will gradually acquire the skills needed for behavior changes while absorbing health information and enhancing their motivation for execution. When the information reserve, potential motivation and behavioral skills reach a certain degree of maturity, the individual will realize a positive change in healthy behaviors [16]. We can use this model to explore the predictors of healthy behaviors, and assume that all variables contained in IMB model will have the chance to be predictors of poor self-management behaviors among COPD patients (Fig. 1).
Fig. 1.
Hypothesized information-motivation-behavioural skills model of self-management behaviors
The aim of this study was to develop and verify a risk prediction model to predict poor self-management behaviors for newly diagnosed COPD patients based on the IMB model. It is necessary to provide a valuable and clinically applicable prediction tool to screen out poor self-management behaviors for newly diagnosed COPD patients. So that clinical nurses can identify high-risk groups at an early stage and compare the risk differences of different COPD patients, and provide a thinking for clinical nurses to formulate appropriate interventions as early as possible.
Methods
Study design and participants
This prospective cohort study was conducted in a general hospital in Chengdu, China. We collected the baseline data from December 2021 to March 2022, and the outcome indicator (self-management behaviors) was evaluated in April 2023. In this study, inclusion criteria were: ①≥40 years old; ②Patients diagnosed with COPD according to the global initiative for chronic obstructive lung disease (2020 REPORT) in two weeks [25]; ③Patients with clear consciousness and normal communication skills, and volunteered to take part in the study. Exclusion criteria in the development cohort and the validation cohort were: individuals combined with dementia and other chronic diseases, such as asthma, asthma and COPD Overlap Syndrome (ACOS), active pulmonary tuberculosis, stroke and COVlD-19.
According to the development requirements of risk prediction model, the participants in the development cohort and the validation cohort need to be different in date or location of inclusion. In this study, we chose hospitalized COPD patients from December 2021 to February 2022 to enter the development cohort, and COPD patients in outpatient service in March 2022 to enter the validation cohort. The follow-up period was one year.
Measures
Self-management behaviors assessment
The COPD Self-Management Scale (CSMS) was used to evaluate self-management behaviors of COPD patients. The CSMS consists of 5 dimensions and 51 items, and the response to each item was from one to five (1 = never, 2 = rarely, 3 = sometimes, 4 = often, 5 = always) [13], and cronbach’s alpha of the scale was 0.822 in this study. The total score of this scale ranged from 51 to 255. The CSMS mainly assessed five domains, including symptom management, daily life management, emotion management, information management, and self-efficacy, and higher scores indicating higher self-management ability. Individuals were categorized as having low ability (scores from 0 to < 144), medium ability (scores from 144 to < 160), and high ability (scores ≥ 160) [26]. We classified participants with low and medium ability into the same type and named them as poor self-management behaviors.
Cognitive function
Cognitive function assessment was conducted by standardised trained and certified researchers after excluding patients with dementia. The Beijing version of Montreal Cognitive Assessment (MoCA) was used to assess global cognitive function after excluding participants with dementia [27]. MoCA includes eight cognitive domains: orientation, language, working memory, concentration, short-term memory, attention, executive function, and visuospatial ability, and is sensitive to evaluate mild cognitive impairment (MCI). MoCA was recommended by “consensus of Chinese memory physical examination experts”, which was used to assess global cognitive function in this study. Taking into account the influence of education level on cognitive function, the MoCA score can be adjusted based on years of education, with one additional point given for individuals with less than 12 years of education. The MoCA has a maximum score of 30, where higher scores indicate better cognitive function, and a score below 26 suggests MCI [27]. The global cognitive function was assessed through face-to-face interviews, lasting approximately 10 min per participant.
Anxiety and depression
Self-Rating Anxiety Scale (SAS) and Self-Rating Depression Scale (SDS) developed by ZUNG et al. [28, 29], were used to assess anxiety and depression. SAS consists of 20 items, using Likert 4-level scoring method, in which score1 to 4 means “no or little time” to “most or all the time” respectively. The final score of SAS was the original score (the sum of the scores of each item) multiplied by 1.25, then taking the integer part. According to the standard of 1158 normal people in China, the patient was anxiety when the score was ≥ 50 [30]; The number of items and scoring method of SDS were consistent with SAS, but the patient was depression when the total score was ≥ 53 according to the standard of 1340 normal people in China [30]. Cronbach’s alpha of SAS and SDS were 0.905 and 0.937 respectively in this study.
Patient activation
Patient Activation Measure (PAM) was used to evaluate patient activation. PAM was first developed by Hibbard in 2004 [24]. Later, Shiqiao Chen translated it into Chinese [31], and cronbach’s alpha in this study was 0.863. PAM was developed to measure a person’s knowledge, confidence, skills, and responsibility to manage diseases. The 13-item PAM version (PAM-13) was used in this study. PAM-13 has 13 items in total, and the answer of each item was from “Disagree strongly” to “Agree strongly”. According to the unique calculation method, their answers are counted and given a total score between 0 and 100, which is converted into four levels. The higher the score, the higher the degree of activation. PAM1(scores<47.0): participants were at a loss for self-management; PAM2(scores 47.1 ∼ 55.1): participants were aware of the importance of self-management, but lacked knowledge and confidence about self-management; PAM3(scores 55.2 ∼ 67): participants could take action to manage their health, but they still lacked enough confidence; PAM4(scores ≥ 67.1): participants had a good ability to manage their health and mastered some skills of self-management [24].
Social support
The Social Support Rate Scale (SSRS) was used to evaluate the participants’ social support level, which was created by Xiao in 1993 [32]. The SSRS included the physical and mental support from family and society. The content included objective social support, subjective social support and social support utilization, with a total of 10 projects. The total score of the scale is 66, and the higher the score, the higher the social support level. A total score of 45–66 indicated a high level of social support, 23–44 indicated a moderate level, and less than 22 indicated a low level of social support [32]. Cronbach’s alpha of SSRS was 0.916 in this study.
Frailty
Frailty is a state of increased vulnerability and decreased resistance to stress in the organism, resulting from aging or various health issues [33]. Frailty assessment scale for COPD patients was used to evaluate the frailty of COPD patients. This scale was developed by Yang et al. in 2021, and which contained 5 dimensions (symptoms, physiological status, health information, psychological mood and social adaptation) and 24 items, and higher scores indicating more frailty [33]. Cronbach’s alpha of this scale was 0.742.
Data collection
Researchers reviewed the medical records and identified participants who met the inclusion criteria. The researchers explained the study in detail to patients who met the criteria and agreed to provide written informed consent. The staff assisted patients who needed help to answer the questionnaires. Demographic and disease-related data were collected on gender, age, education level, occupation, marital status, living arrangement, habitual residence, family monthly income, financial burden, body mass index (BMI), nutritional assessment results, smoking or not (participants who never smoked refer to participants who never smoked or had smoked < 100 cigarettes in their lives; Participants who smoke refer to participants who smoked at least one cigarette every day now or ever ) [34], current drinker (defined as a person who drinks at least one glass of wine per week and lasted for 1 year and above). Comorbidity, physical activity (including walking, exercise, house cleaning, and measured by the question “did you get more than 30 minutes of physical activity a day”). These characteristics were self-reported by the participants based on the questionnaire. The researchers also collected disease-related information from the medical records. For example, lung function indicators “the percentage of predicted FEV1 (FEV1% pred)”. The lung function was divided into four degrees from mild to extremely severe according to FEV1%pred. GOLD1: FEV1%pred ≥ 80%; GOLD2: 50% ≤ FEV1%pred < 80%; GOLD3: 30% ≤ FEV1%pred < 50%; GOLD4: FEV1%pred < 30%). The degree of dyspnea was evaluated with the mMRC, which was rated from level 0 (not troubled with breathlessness, except with strenuous exercise) to level 4 (too breathless to leave the house or breathless when dressing or undressing [25].
After collecting the above information except self-management behaviors at baseline, researchers explained the knowledge of self-management to each participant in detail, including the identification and emergency treatment of acute disease, the method of respiratory function exercise, the use of inhalants and other drugs, the basic points for attention in oxygen therapy and the related guidance for mood adjustment. In addition, all participants did not receive any extra intervention or guidance. We evaluated self-management behaviors of each participant one year later by the COPD Self-Management Scale, which can be found in part “self-management behaviors assessment” above.
Ethics statement
The study was approved by the Bioethics Committee of Medical Center Hospital of Qionglai, China (NO, 202203). All participants understood the aims and methods of the study, and they all agreed to participate in the study and signed the informed consent form.
Statistical analysis
We used R soft-ware (version 4.1.0; R Foundation for Statistical Computing, Vienna, Austria) and SPSS Statistics 23.0 (IBM Corp, Armonk, NY, USA) for data analysis. Continuous variables obeying normal distribution were described as mean ± SD. Continuous variables that did not obey normal distribution were expressed as median (quartiles) [M (P25, P75)], and categorical variables as frequencies and percentages. Using ANOVA for continuous variables and the chi-square test for categorical variables to determine the differences between the development cohort and the validation cohort. We used SPSS Statistics 23.0 (IBM Corp, Armonk, NY, USA) to develop the model in the development cohort. Firstly, univariate logistic regression analysis was used to screen out the significant risk factors of poor self-management behaviors for COPD patients. All variables at the level of P < 0.05 were entered to multivariate logistic regression analysis in order to identify the final predictive factors of poor self-management behaviors for COPD patients, and the variables with P < 0.05 in multivariate logistic regression analysis were entered in the final prediction model, and we used the R language rms package to construct a nomogram was according to this result.
We used the R language pROC package for receiver characteristic curve (ROC) operation in the development and validation cohort respectively. The area under curve (AUC) was calculated to test the performance of nomogram in the development and validation cohort. An AUC less than 0.65 indicates poor model discrimination, 0.65 to 0.75 indicates that the model has some discriminatory ability, and a value greater than 0.75 indicates that the model has a good discriminatory ability [35]. We used the R language rms package for calibration curve operation to evaluate the calibration of nomogram, accompanied by the Hosmer-Lemeshow test (P > 0.05 indicates a good fit). Moreover, the nomogram was subjected to 1000 bootstrap resamples for internal validation in the development cohort, and for external validation in the verification cohort. Finally, we used the R language nricens package for clinical decision curves operation to assess the clinical benefit of the nomogram [36].
Results
The characteristics of participants
Table 1 showed the characteristics of participants from both cohorts. During one-year follow-up period, the incidence of poor self-management behaviors and MCI were 68.3% and 58% respectively in total participants (the data of the two cohorts were similar). 26.6% of the participants had a tendency to be depression, 45.6% of the participants were at low activation level (PAM1 + PAM2), while 39.3% were at high activation level (PAM4). The majority of the participants (71.9%) were made up of GOLD 1 and GOLD 2. There was no significant difference in characteristics between the two cohorts (P > 0.05) (Table 1).
Table 1.
Characteristics of patients with COPD in development and validation cohorts
| Descriptive variables | Total (N = 331) |
Development cohort (N = 222) |
Validation cohort (N = 109) | P value |
|---|---|---|---|---|
| Gender, N (%) | 0.307 | |||
| Male | 212(64.0) | 138(62.2) | 74(67.9) | |
| Female | 119(36.0) | 84(37.8) | 35(32.1) | |
| Age (yrs), N (%) | 0.921 | |||
| 40 ~ 64 | 50(15.1) | 32(14.4) | 18(16.5) | |
| 65 ~ 74 | 140(42.3) | 97(43.7) | 43(39.4) | |
| 75+ | 141(42.6) | 93(41.9) | 48(44.0) | |
| Education level, N (%) | 0.908 | |||
| Junior middle school and below | 214(64.7) | 144(64.9) | 70(64.2) | |
| High school and above | 117(35.3) | 78(35.1) | 39(35.8) | |
| Occupation, N (%) | 0.169 | |||
| Farm laborer | 222(67.1) | 140(63.1) | 82(75.2) | |
| Worker | 56(16.9) | 43(19.4) | 13(11.9) | |
| Freelance work | 22(6.6) | 16(7.2) | 6(5.5) | |
| Civil servants and professional | 31(9.4) | 23(10.4) | 8(7.3) | |
| Marital status, N (%) | 0.428 | |||
| Married | 240(72.5) | 159(71.6) | 81(74.3) | |
| Divorced | 12(3.6) | 6(2.7) | 6(5.5) | |
| Widowed | 67(20.2) | 49(22.1) | 18(16.5) | |
| Single | 12(3.6) | 8(3.6) | 4(3.7) | |
| Living arrangement, N (%) | 0.690 | |||
| Living alone | 16(4.8) | 10(4.5) | 6(5.5) | |
| Not live alone | 315(95.2) | 212(95.5) | 103(94.5) | |
| Frequent dwelling place, N (%) | 0.103 | |||
| Rural | 201(60.7) | 128(57.7) | 73(67.0) | |
| Urban | 130(39.3) | 94(42.3) | 36(33.0) | |
| Monthly household income, N (%) | 0.352 | |||
| <5000 (¥) | 61(18.4) | 44(19.8) | 17(15.6) | |
| ≥5000 (¥) | 270(81.6) | 178(80.2) | 92(84.4) | |
| Financial burden, N (%) | 0.827 | |||
| No | 219(66.2) | 146(65.8) | 73(67.0) | |
| Yes | 112(33.8) | 76(34.2) | 36(33.0) | |
| Smoking or not, N (%) | 0.693 | |||
| No | 160(48.3) | 109(49.1) | 51(46.8) | |
| Yes | 171(51.7) | 113(50.9) | 58(53.2) | |
| Current drinker, N (%) | 0.593 | |||
| No | 243(73.4) | 165(74.3) | 78(71.6) | |
| Yes | 88(26.6) | 57(25.7) | 31(28.4) | |
| Physical activity, N (%) | 0.712 | |||
| No | 193(58.3) | 131(59.0) | 62(56.9) | |
| Yes | 138(41.7) | 91(41.0) | 47(43.1) | |
| MoCA(median; IQR) | 21(18 ~ 27) | 21(18 ~ 27) | 22(18.5 ~ 26) | 0.592 |
| Mild cognitive impairment, N (%) | 0.598 | |||
| No | 139(42.0) | 91(41.0) | 48(44.0) | |
| Yes | 192(58.0) | 131(59.0) | 61(56.0) | |
| Anxiety, N (%) | 0.837 | |||
| No | 210(63.4) | 140(63.1) | 70(64.2) | |
| Yes | 121(36.6) | 82(36.9) | 39(35.8) | |
| Depression, N (%) | 0.593 | |||
| No | 243(73.4) | 165(74.3) | 78(71.6) | |
| Yes | 88(26.6) | 57(25.7) | 31(28.4) | |
| Malnutrition, N (%) | 0.383 | |||
| No | 242(73.1) | 159(71.6) | 83(76.1) | |
| Yes | 89(26.9) | 63(28.4) | 26(23.9) | |
| Comorbidity, N (%) | 0.088 | |||
| No | 166(50.2) | 105(47.3) | 61(56.0) | |
| One | 90(27.2) | 61(27.5) | 29(26.6) | |
| Two or more | 75(22.6) | 56(25.2) | 19(17.4) | |
| BMI(kg/m2), N (%) | 0.998 | |||
| <18.5 | 32(9.7) | 23(10.4) | 9(8.3) | |
| 18.5 ~ 24.9 | 208(62.8) | 137(61.7) | 71(65.1) | |
| 25-27.9 | 57(17.2) | 38(17.1) | 19(17.4) | |
| ≥28 | 34(10.3) | 24(10.8) | 10(9.2) | |
| FEV1%pred, N (%) | 0.057 | |||
| GOLD1 | 97(29.3) | 60(27.0) | 37(33.9) | |
| GOLD2 | 141(42.6) | 92(41.4) | 49(45.0) | |
| GOLD3 | 88(26.6) | 67(30.2) | 21(19.3) | |
| GOLD4 | 5(1.5) | 3(1.4) | 2(1.8) | |
| mMRC, N (%) | 0.357 | |||
| Level 0 + Level 1 + Level 2 | 131(39.6) | 84(37.8) | 47(43.1) | |
| Level 3 + Level 4 | 200(60.4) | 138(62.2) | 62(56.9) | |
| Frailty (median; IQR) | 12(10 ~ 16) | 12(9 ~ 16) | 12(10 ~ 16) | 0.427 |
| Social support, N (%) | 0.578 | |||
| Low | 81(24.5) | 55(24.8) | 26(23.9) | |
| Medium | 232(70.1) | 157(70.7) | 75(68.8) | |
| High | 18(5.4) | 10(4.5) | 8(7.3) | |
| PAM, N (%) | 0.631 | |||
| level 1 + level 2 | 151(45.6) | 102(45.9) | 49(44.9) | |
| level 3 | 50(15.1) | 28(12.6) | 22(20.2) | |
| level 4 | 130(39.3) | 92(41.4) | 38(34.9) | |
| Disorder of self-management, N (%) | 0.390 | |||
| No | 105(31.7) | 67(30.2) | 38(34.9) | |
| Yes | 226(68.3) | 155(69.8) | 71(65.1) |
BMI, Body Mass Index; FEV1%pred, The percentage of predicted FEV1; mMRC, dyspnoea rating scale proposed by the British Medical Research Council; PAM, Patient Activation Measure
Development of the prediction model
In the development cohort, the univariate logistic regression analysis showed that factors such as age, education level, occupation, frequent dwelling place, financial burden, physical activity, MoCA, depression, anxiety, comorbidity, FEV1%pred, mMRC, frailty, social support and PAM were significantly related to poor self-management behaviors (all P < 0.05) (Table 2, the univariate regression analysis). Putting the above variables into the multivariate logistic regression model, the results showed that MoCA (95%CI 0.453 ∼ 0.731), depression (95%CI 1.135 ∼ 39.241), PAM3 (95%CI 0.006 ∼ 0.265), PAM4 (95%CI 0.011 ∼ 0.234) were association with poor self-management behaviors for COPD patients (all P < 0.05) (Table 2, the multivariate regression analysis). After fitting the above three variables again in the multivariate regression analysis, the results showed that depression (AOR = 10.838, 95%CI 2.929 ∼ 40.110, P < 0.001), MoCA (AOR = 0.591, 95%CI 0.505 ∼ 0.692, P < 0.001), PAM3 (AOR = 0.042, 95%CI 0.010 ∼ 0.180, P < 0.001), PAM4 (AOR = 0.038, 95%CI 0.012 ∼ 0.121, P < 0.001) were the ultimate risk predictors of poor self-management behaviors for COPD patients (Table 3).
Table 2.
Logistic regression in the development cohort for poor self-management behaviors
| Descriptive variables | Univariate regression analysis | Multivariate regression analysis | ||
|---|---|---|---|---|
| OR (95% CI) | P value | AOR (95% CI) | P value | |
| Gender(ref = Male) | ||||
| Female | 1.032(0.571,1.866) | 0.916 | NA | |
| Age (yrs)(ref = 40 ~ 64) | ||||
| 65 ~ 74 | 5.159(2.173, 12.248) | < 0.001 | 1.290(0.213,7.812) | 0.782 |
| 75+ | 10.587(4.214,26.601) | < 0.001 | 1.095(0.159,7.560) | 0.927 |
| Education level(ref = Junior middle school and below) | ||||
| High school and above | 0.095(0.049,0.185) | < 0.001 | 1.497(0.407,5.508) | 0.544 |
| Occupation(ref = Farm laborer) | ||||
| Worker | 0.214(0.101,0.455) | < 0.001 | 0.626(0.115,3.409) | 0.588 |
| Freelance work | 0.311(0.102,0.943) | < 0.001 | 6.971(0.588,82.681) | 0.124 |
| Civil servants and professional | 0.039(0.012,0.127) | < 0.001 | 0.870(0.068,11.197) | 0.915 |
| Marital status(ref = Married) | ||||
| Divorced | 0.229(0.041,1.294) | 0.095 | NA | |
| Widowed | 1.585(0.749,3.354) | 0.229 | NA | |
| Single | 1.376(0.268,7.058) | 0.702 | NA | |
| Living arrangement(ref = Living alone) | ||||
| Not live alone | 1.577(0.430,5.780) | 0.492 | NA | |
| Frequent dwelling place(ref = Rural) | ||||
| Urban | 0.205(0.110,0.380) | < 0.001 | 0.380(0.072,1.998) | 0.253 |
| Monthly household income(ref=<5000) | ||||
| ≥5000 (¥) | 0.448(0.196,1.025) | 0.057 | NA | |
| Financial burden(ref = No) | ||||
| Yes | 2.884(1.434,5.643) | 0.003 | 0.610(0.161,2.315) | 0.467 |
| Smokers(ref = No) | ||||
| Yes | 1.197(0.675,2.125) | 0.539 | NA | |
| Current drinker(ref = No) | ||||
| Yes | 0.821(0.430,1.564) | 0.548 | NA | |
| Physical activity(ref = No) | ||||
| Yes | 0.275(0.251,0.501) | < 0.001 | 0.417(0.113,1.539) | 0.189 |
| MoCA | 0.684(0.617,0.760) | < 0.001 | 0.576(0.453,0.731) | < 0.001 |
| Anxiety(ref = No) | ||||
| Yes | 2.639(1.369,5.089) | 0.004 | 0.684(0.199,2.350) | 0.546 |
| Depression(ref = No) | ||||
| Yes | 4.082(1.741,9.570) | 0.001 | 6.675(1.135,39.241) | 0.036 |
| Malnutrition(ref = No) | ||||
| Yes | 1.978(0.989,3.955) | 0.054 | NA | |
| Comorbidity(ref = No) | ||||
| One | 0.466(0.238,0.911) | 0.026 | 1.283(0.309,5.328) | 0.731 |
| Two or more | 1.038(0.492,2.191) | 0.921 | 0.676(0.175,2.613) | 0.570 |
| BMI(kg/m2)(ref=<18.5) | ||||
| 18.5 ~ 24.9 | 0.460(0.148,1.435) | 0.181 | NA | |
| 25-27.9 | 0.405(0.114,1.441) | 0.163 | NA | |
| ≥28 | 0.511(0.127,2.057) | 0.345 | NA | |
| FEV1%pred(ref = GOLD1) | ||||
| GOLD2 | 3.463(1.738,6.900) | < 0.001 | 1.254(0.342,4.592) | 0.733 |
| GOLD3 | 7.877(3.310,18.745) | < 0.001 | 1.842(0.240,14.145) | 0.557 |
| GOLD4 | 2.444(0.210,28.433) | 0.475 | 5.864(1.006,60.720) | 0.618 |
| mMRC(ref = Level 0 + Level 1 + Level 2) | ||||
| Level 3 + Level 4 | 8.351(4.364, 15.980) | < 0.001 | 2.188(0.627,7.629) | 0.219 |
| Frailty | 1.272(1.170,1.383) | < 0.001 | 1.033(0.859,1.241) | 0.732 |
| Social support(ref = Low) | ||||
| Medium | 0.568(0.270,1.193) | 0.135 | 0.210(0.046,1.960) | 0.054 |
| High | 0.063(0.012,0.337) | 0.001 | 0.087(0.003,2.322) | 0.145 |
| PAM(ref = level 1 + level 2) | ||||
| level 3 | 0.133(0.045,0.394) | < 0.001 | 0.038(0.006,0.265) | 0.001 |
| level 4 | 0.062(0.026,0.148) | < 0.001 | 0.050(0.011,0.234) | < 0.001 |
Variables with p value less than 0.05 in univariate logistic regression analysis were subjected to multivariate logistic regression analysis. and when dealing with a categorical variable that has 3 or more levels, if any of these levels were found to be significant in the univariate analysis, it was imperative to include all levels in the subsequent multivariate regression analysis
The included variables were age, education level, occupation, frequent dwelling place, financial burden, physical activity, MoCA, depression, anxiety, comorbidity, FEV1%pred, mMRC, frailty, social support and PAM in the multivariate regression analysis of this study
BMI, Body Mass Index; FEV1%pred, The percentage of predicted FEV1; mMRC, dyspnoea rating scale proposed by the British Medical Research Council
PAM, Patient Activation Measure; NA, not entry; AOR, adjusted OR
Table 3.
Multivariate regression analysis of the three predictive factors were selected from Table 2
| Descriptive variables | β | AOR (95% CI) | P value |
|---|---|---|---|
| Depression | 2.383 | 10.838(2.929,40.110) | < 0.001 |
| MoCA | -0.526 | 0.591(0.505,0.692) | < 0.001 |
| PAM: level 3 | -3.179 | 0.042(0.010,0.180) | < 0.001 |
| PAM: level 4 | -3.262 | 0.038(0.012,0.121) | < 0.001 |
Development of risk prediction nomogram for poor self-management behaviors
The nomogram for poor self-management behaviors was developed in accordance with the multivariate regression analysis in Table 3. According to the nomogram (Fig. 2), when the patient had depressive symptoms, the score was 40 points. When the patient was in the low activation level, the score was 57.5, and the worse the cognitive function, the higher the score. If a total score was above 60, the probability of poor self-management behaviors was greater than 50%. The AUC of the model was 0.945 (95% CI: 0.911 ∼ 0.980), as shown in Fig. 3A. At the maximum Youden index (0.807), the sensitivity and specificity were 89.68% and 91.04% respectively. The calibration plot for poor self-management behaviors indicated an optimal agreement between the predictions through the nomogram and actual observations. After the Hosmer-Lemeshow test, the results indicated good model prediction (χ2 = 13.873, P = 0.085) (Fig. 4A).
Fig. 2.
Nomogram of risk prediction of behavioral disorder of self-management for newly diagnosed COPD patients
Fig. 3.
ROC curves of the predictive model for risk (A: The AUC and its 95% CI were 0.945(0.911 ∼ 0.980) in development cohort; B: The AUC and its 95% CI were 0.898(0.831 ∼ 0.965) in validation cohort)
Fig. 4.
Calibration curve of the nomogram model (A: Calibration curve of internal validation in development cohort (Hosmer-Lemeshow test χ2 = 13.873, P = 0.085); B: Calibration curve of external validation in validation cohort(Hosmer-Lemeshow test χ2 = 9.722, P = 0.285). The diagonal dotted line showed an ideal model for the perfect prediction ability. the Apparent line represents the reality performance of the nomogram. and the bias-corrected line, the closer fit to the diagonal dotted line, the better prediction ability of the nomogram
Validation of predictive accuracy of the nomogram for poor self-management behaviors
The incidence of poor self-management behaviors in the validation cohort was 65.1% (Table 1). The nomogram indicated good discriminative ability with an AUC of 0.898 (95% CI 0.831 ∼ 0.965) (Fig. 3B). The calibration curve indicated a good agreement between prediction and observation in the probability of behavioral disorder of self-management, the results of Hosmer-Lemeshow test indicated good model prediction (χ2 = 9.722, P = 0.285) (Fig. 4B).
Discussion
This is the first study to develop and verify a risk prediction model to predict poor self-management behaviors for newly diagnosed COPD patients. We considered the related factors that affect self-management behaviors according to the IMB model. Three risk predictors inclued in our risk prediction model, namely cognitive function, depression and patient activation. The model could better predict poor self-management behaviors for newly diagnosed COPD patients, and showed good calibration in the external validation cohort. The evaluation process of the three variables is relatively simple, and the total number of items contained in the three scales is also less than CSMS. In addition, some self-management behaviors for newly diagnosed COPD patients may not be judged by the CSMS in a short time, such as the ability to deal with acute attacks, and the choice of drugs during the attack. Therefore, it is of great significance to establish the risk prediction model to predict poor self-management behaviors for newly diagnosed COPD patients.
In this study, the incidence of poor self-management behaviors for COPD patients was 68.3% in total (69.8% in the development cohort, and 65.1% in the validation cohort). But our study was different from a study in Chongqing,.which showed the poor self-management behaviors for newly diagnosed COPD patients was 50% higher than other COPD patients due to mild symptoms, or not knowing enough about the disease [15]. On the other hand, the majority of the newly diagnosed COPD patients had mild or moderate lung function impairment [15], which was similar to the population included in our study (68.4% and 78.9% ), but most individuals in our study had obvious dyspnea (62.2% and 56.9%). During this study, COVID−19 was in the popular period in China. In order to exclude the influence of COIVD−19 on the research results, we excluded COPD patients with COVID−19 when selecting participants. The severity of pulmonary function damage in COPD patients was not necessarily positively correlated with the severity of dyspnea. Therefore, when COPD patients had mild or moderate pulmonary function impairment, they might still have severe dyspnea [37]. It may also be that the information about dyspnea came from medical records, and the patient was in an acute attack state when doctors collected the information, therefore, there were many patients with severe dyspnea in our study.
A previous study showed that higher income and education level, younger age had a positive impact on self-management behaviors for COPD patients [26]. Another study found that the lung function and dyspnea were also related to self-management behaviors [38]. The univariate regression analysis of this study also showed the same results, but the prediction model constructed by the results of multivariate regression analysis only included three predictive factors (cognitive function, patient activation and depression). It might be that there were some potential collinearity relationships between some variables. Researches showed that education and age were potentially related to cognitive function. A study reported that 60% of people over 60 years old would have cognitive impairment because of the gradual decline of brain quality [39], and another study confirmed the decisive role of early education and lifelong education in preventing and delaying cognitive decline [40]. This study might indicate that there was a potential relationship among cognitive function, age and education level, and cognitive function had greater impact on self-management behaviors. Cognitive impairment in COPD patients was common due to neuron injury, chronic hypoxemia, systemic inflammation, vascular-mediated brain injury, alterations in cerebral perfusion, or a reduction in gray matter volume brought on by chronic hypoxia [41]. In this study, the incidence of MCI was about 58%, cognitive impairment for COPD patients was mainly manifested in slow information processing, delayed memory and attention deficit, which were the main factors that determine self-management behaviours [41]. This might be one of the main reasons why cognitive function could be used as a predictor of self-management behaviours [42].
In addition, predictors of this study also included patient activation and depression. Some studies believed that patient activation played a vital role on self-management behaviors [24, 43]. Patient activation, which was an effective index to measure patient activation of chronic patients, could evaluate patients’ knowledge, skills and confidence on self-management behaviors [43]. Hibbard, who was the developer of PAM, declared that patients with low scores were passive nursing recipients, and they didn’t think it was necessary to play an active role as patients, but those with high scores were more active in their own health and could adhere to self-management [24]. A study showed that patient activation can improve the participation in self-care practice and behavioral change plan of healthy lifestyle [38], which was similar to our results. Our study indicated that PAM 3(OR = 0.042, 95%CI 0.010 ∼ 0.180, P < 0.001) and PAM 4(OR = 0.038, 95%CI 0.012 ∼ 0.121, P < 0.001) were protective factors on poor self-management behaviors.
Depression was another predictor on poor self-management behaviors in COPD patients (OR = 10.838, 95%CI 2.929 ∼ 40.110, P < 0.001), which was consistent with Jassem et al. [44]. It may be difficult to motivate patients with depression to participate in self-management plan. Compared with non-depressed COPD patients, COPD patients with depression were more difficult to maintain self-management motivation, so patients with depression were prone to adverse health outcomes [3]. However, physical activity and social support had positive impact on self-management in univariate analysis in this study, but not included in the results of multivariate analysis. This may be due to the collinear relationship between these factors and patient activation or depression. In addition, newly diagnosed COPD patients lacked disease-related knowledge, especially the patients in the outpatient clinics. Moreover, without taking into account lifestyle changes like food and exercise, doctors mostly focused on stopping smoking, becoming worse, and medicine, and medical professionals in outpatient department also lacked the time to fully explain pertinent information to patients [45]. Therefore, newly diagnosed COPD patients might be prone to depression due to fear of the future.
The discrimination of model was evaluated by the area under the ROC curve (AUC). The AUC in our prediction model was 0.945 (95%CI:0.911 ∼ 0.980). It is generally believed that a value of AUC greater than 0.75 indicates that the model has a good discriminatory. This indicated that our prediction model had good discrimination ability. The AUC of the validation cohort was 0.898(95%CI:0.831 ∼ 0.965), which showed that the predictive model was still good in discrimination after verification. Hosmer-Lemeshow test and calibration curve were used to test the goodness of fit of the prediction model. The results of Hosmer-Lemeshow test of this model (χ2 = 13.873, P = 0.085) showed a good degree of fit of this model. In addition, at the maximum Youden index (0.807), the sensitivity and specificity were 89.68% and 91.04% respectively, which showed that this model could accurately identify patients with poor self-management behaviours and also could precisely exclude false positive patients. In addition, high sensitivity of prediction tool has important clinical significance, and effectively reduce the incidence of poor self-management behaviors among the newly diagnosed COPD patients. Moreover, this study constructed a nomogram based on logistic regression analysis. The nomogram depended on the user-friendly digital interface, which improved the accuracy and provided a more understandable prognosis to help better clinical decision-making [46]. For example, The three predictive factors (cognitive function, patient activation, and depression) in this model would be assigned a score respectively based on the nomogram, and calculated the total score of the three predictive factors (if a total score was above 60, the probability of poor self-management behaviors was greater than 50%). Then according to the predicted results, nurses could pay more attention to patients with high risk of poor self-management behaviors and intervene to them as early as possible. This was the first study to use nomogram to predict the risk of poor self-management behaviours in newly diagnosed COPD patients, and risk predictors included were easy to obtain and evaluate (not involving biochemical and other physical examination indicators), which were not only suitable for inpatients, but also suitable for outpatients with COPD. However, the value of the three predictive factors included in the model on self-management behaviors is still unclear, and we may need further study to compare their value on self-management behaviors for COPD patients.
There were some limitations in this study. Firstly, the follow-up time of this study was one year, patients need to recall some contents when filling in the CSMS, which may cause bias in results; Moreover, this risk prediction model was mainly considered to newly diagnosed COPD patients, the course of disease and long-term family oxygen therapy were not included in, which may lead to incomplete in results. In addition, although a large number of influencing factors have been considered in our study, there may still be potential confounding factors that have not been considered.
Conclusion
The purpose of this prospective cohort study was to develop and externally verify a risk prediction model on poor self-management behaviors for newly diagnosed COPD patients. The final risk prediction model included three predictive factors (cognitive function, depression and patient activation). We suggest evaluating the risk of poor self-management behaviors for newly diagnosed COPD patients in the outpatient department or the inpatient department, and to identify high risk groups of poor self-management behaviors as early as possible, so that medical care providers can take preventive measures.
Acknowledgements
Not applicable.
Abbreviations
- COPD
Chronic Obstructive Pulmonary Disease
- CSMS
COPD Self-Management Scale
- MoCA
Montreal Cognitive Assessment
- PAM
Patient Activation Measure
- SAS
Self-Rating Anxiety Scale
- SDS
Self-Rating Depression Scale
- SSRS
Social Support Rate Scale
- FEV1% pred
The Percentage of Predicted FEV1
- BMI
Body Mass Index
- mMRC
the modified Medical Research Council Dyspnea Scale
Author contributions
C contributed to the study design, performed the statistical analyses, interpreted the data and drafted the manuscript. L, H, W, D, Z, S contributed to the statistical analyses and interpretation of the data and critically revised the manuscript. Y designed and supervised the study, interpreted the data, and critically revised the manuscript. All authors read and approved the final manuscript.
Funding
This study was supported by National Social Science Funding Project (20BRK039).
Data availability
Data is provided within the manuscript or supplementary information files.
Declarations
Ethics approval and consent to participate
The study was approved by the Bioethics Committee of Medical Center Hospital of Qionglai, China (NO,202203). The study was initiated after the ethical approval. Informed written consent was obtained from the voluntary study participants in Chinese. Anonymity and confidentiality were maintained all through the study. Our study adhered to the declaration of Helsinki.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Ruvuna L, Sood A. Epidemiology of chronic obstructive pulmonary disease. Clin Chest Med. 2020;41:315–27. [DOI] [PubMed] [Google Scholar]
- 2.Fang L, Gao P, Bao H, Tang X, Wang B, Feng Y, et al. Chronic obstructive pulmonary disease in China: a nationwide prevalence study. Lancet Respir Med. 2018;6(6):421–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Mannino DM, Roberts MH, Mapel DW, Zhang Q, Lunacsek O, Grabich S, et al. National and local direct medical cost burden of COPD in the united States from 2016 to 2019 and projections through 2029. Chest. 2024;165(5):1093–106. [DOI] [PubMed] [Google Scholar]
- 4.Kitamura S, Igarashi A, Yamauchi Y, Senjyu H, Horie T, Yamamoto-Mitani N. Self-management activities of older people with chronic obstructive pulmonary disease by types of healthcare services utilised: A cross-sectional questionnaire study. Int J Older People Nurs. 2020;15(3):e12316. [DOI] [PubMed] [Google Scholar]
- 5.Wang LH, Zhao Y, Chen LY, Zhang L, Zhang YM. The effect of a nurse-led self-management program on outcomes of patients with chronic obstructive pulmonary disease. Clin Respir J. 2020;14(2):148–57. [DOI] [PubMed] [Google Scholar]
- 6.Effing TW, Vercoulen JH, Bourbeau J, Trappenburg J, Lenferink A, Cafarella P, et al. Definition of a COPD self-management intervention: international expert group consensus. Eur Respir J. 2016;48(1):46–54. [DOI] [PubMed] [Google Scholar]
- 7.Putcha N, Drummond MB, Wise RA, Hansel NN. Comorbidities and chronic obstructive pulmonary disease: prevalence, influence on outcomes, and management. Semin Respir Crit Care Med. 2015;36(4):575–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Lan X, Lu X, Yi B, Chen X, Jin S. Factors associated with self-management behaviors of patients with chronic obstructive pulmonary disease. Jpn J Nurs Sci. 2022;19(1):e12450. [DOI] [PubMed] [Google Scholar]
- 9.Yang H, Wang H, Du L, Wang Y, Wang X, Zhang R. Disease knowledge and self-management behavior of COPD patients in China. Med (Baltim). 2019;98(8):e14460. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.O’Conor R, Muellers K, Arvanitis M, Vicencio DP, Wolf MS, Wisnivesky JP, et al. Effects of health literacy and cognitive abilities on COPD self-management behaviors: A prospective cohort study. Respir Med. 2019;160:105630. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Spruit MA, Pitta F, McAuley E, ZuWallack RL, Nici L. Pulmonary rehabilitation and physical activity in patients with chronic obstructive pulmonary disease. Am J Respir Crit Care Med. 2015;192(8):924–33. [DOI] [PubMed] [Google Scholar]
- 12.van Beerendonk I, Mesters I, Mudde AN, Tan TD. Assessment of the inhalation technique in outpatients with asthma or chronic obstructive pulmonary disease using a metered-dose inhaler or dry powder device. J Asthma. 1998;35(3):273–9. [DOI] [PubMed] [Google Scholar]
- 13.Zhang C, Wang W, Li J, Cai X, Zhang H, Wang H, et al. Development and validation of a COPD self-management scale. Respir Care. 2013;58(11):1931–6. [DOI] [PubMed] [Google Scholar]
- 14.Lim KE, Kim SR, Kim HY, Kim SR, Lee YC. Self-management model based on information-motivation-behavioral skills model in patients with chronic obstructive pulmonary disease. J Adv Nurs. 2022;78(12):4092–103. [DOI] [PubMed] [Google Scholar]
- 15.Zhang M, Tang T, Wan M, Zhang Q, Wang C, Ma Q. Self-reported reasons for treatment nonadherence in chronic obstructive pulmonary disease (COPD) patients: a 24-week prospective cohort study in China. Ann Palliat Med. 2020;9(5):3495–505. [DOI] [PubMed] [Google Scholar]
- 16.Kim CJ, Kang HS, Kim JS, Won YY, Schlenk EA. Predicting physical activity and cardiovascular risk and quality of life in adults with osteoarthritis at risk for metabolic syndrome: A test of the information-motivation-behavioral skills model. Nurs Open. 2020;7(4):1239–48. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Peters JB, Antons JC, Koolen EH, van Helvoort HAC, van Hees HWH, van den Borst B, et al. Patient activation is a treatable trait in patients with chronic airway diseases: an observational study. Front Psychol. 2022;13:947402. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Okonkwo O, Griffith HR, Belue K, Lanza S, Zamrini EY, Harrell LE, et al. Medical decision-making capacity in patients with mild cognitive impairment. Neurology. 2007;69(15):1528–35. [DOI] [PubMed] [Google Scholar]
- 19.Lid D, Huang JT, Yao WX. Study on correlation between social support and self-management ability in patients with cancerrelated fatiguely. Chin Nurs Res. 2020;34(2):299–302. (Chinese). [Google Scholar]
- 20.Zhao D, Yu L. Influence of social interaction on cognitive functions in the elderly. Adv Psychol Sci. 2016;24(1):46–54. (Chinese). [Google Scholar]
- 21.Luo J, Zhang D, Tang W, Dou LY, Sun Y, et al. Impact of frailty on the risk of exacerbations and All-Cause mortality in elderly patients with stable chronic obstructive pulmonary disease. Clin Interv Aging. 2021;16:593–601. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Dent E, Martin FC, Bergman H, Woo J, Romero-Ortuno R, Walston JD. Management of frailty: opportunities, challenges, and future directions. Lancet. 2019;394(10206):1376–86. [DOI] [PubMed] [Google Scholar]
- 23.Cicutto L, Brooks D, Henderson K. Self-care issues from the perspective of individuals with chronic obstructive pulmonary disease. Patient Educ Couns. 2004;55(2):168–76. [DOI] [PubMed] [Google Scholar]
- 24.Hibbard JH, Stockard J, Mahoney ER, Tusler M. Development of the patient activation measure (PAM): conceptualizing and measuring activation in patients and consumers. Health Serv Res. 2004;39(4 Pt 1):1005–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Halpin DMG, Criner GJ, Papi A, Singh D, Anzueto A, Martinez FJ, et al. Global initiative for the diagnosis, management, and prevention of chronic obstructive lung disease. The 2020 GOLD science committee report on COVID-19 and chronic obstructive pulmonary disease. Am J Respir Crit Care Med. 2021;203(1):24–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Wang L, Nygårdh A, Zhao Y, Mårtensson J. Self-management among patients with chronic obstructive pulmonary disease in China and its association with sociodemographic and clinical variables. Appl Nurs Res. 2016;32:61–6. [DOI] [PubMed] [Google Scholar]
- 27.Gong D, Peng Y, Liu X, Zhang J, Deng M, Yang T, et al. Dose health education on dementia prevention have more effects on community residents when a community physician/nurse leads it? A cross-sectional study. Front Public Health. 2023;11:1101913. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.ZUNG WW, A SELF-RATING DEPRESSION. SCALE Arch Gen Psychiatry. 1965;12:63–70. [DOI] [PubMed] [Google Scholar]
- 29.Zung WW. A rating instrument for anxiety disorders. Psychosomatics. 1971;12(6):371–9. [DOI] [PubMed] [Google Scholar]
- 30.Zhang MY. Handbook of psychiatric rating scale (the second edition). Hunan science and technology publishing house. 1998: 35–39.(Chinese).
- 31.Chen SQ, Cheng PX, Li ZX, Zang YL. Reliability and validity analysis for the Chinese version of the patientactivation measure for patients with chronic heart failure. J Nurses Train. 2020;35(03):198–203. (Chinese). [Google Scholar]
- 32.Xiao SY. Theoretical basis and research application of evaluation scale of social support. J Clin Psychiatry. 1994;4(02):98–100. (Chinese). [Google Scholar]
- 33.Yang J, Guo H, Li M, Zhang CH. Development and analysis of the frailty assessment scale for COPD patients. J Hainan Med Univ. 2021;27:1378–83. (Chinese). [Google Scholar]
- 34.Wei S, Wang D, Wei G, Wang J, Zhou H, Xu H, et al. Association of cigarette smoking with cognitive impairment in male patients with chronic schizophrenia. Psychopharmacology. 2020;237(11):3409–16. [DOI] [PubMed] [Google Scholar]
- 35.Iasonos A, Schrag D, Raj GV, Panageas KS. How to build and interpret a nomogram for cancer prognosis. J Clin Oncol. 2008;26(8):1364–70. [DOI] [PubMed] [Google Scholar]
- 36.Kerr KF, Brown MD, Zhu K, Janes H. Assessing the clinical impact of risk prediction models with decision curves: guidance for correct interpretation and appropriate use. J Clin Oncol. 2016;34(21):2534–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Reilly CC, Jolley CJ, Elston C, Moxham J, Rafferty GF. Blunted perception of neural respiratory drive and breathlessness in patients with cystic fibrosis. ERJ Open Res. 2016;2(1):00057–2015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Bos-Touwen I, Schuurmans M, Monninkhof EM, Korpershoek Y, Spruit-Bentvelzen L, Ertugrul-van der Graaf I, et al. Patient and disease characteristics associated with activation for self-management in patients with diabetes, chronic obstructive pulmonary disease, chronic heart failure and chronic renal disease: a cross-sectional survey study. PLoS ONE. 2015;10(5):e0126400. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Andrade C, Radhakrishnan R. The prevention and treatment of cognitive decline and dementia: an overview of recent research on experimental treatments. Indian J Psychiatry. 2009;51(1):12–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Sattler C, Toro P, Schönknecht P, Schröder J. Cognitive activity, education and socioeconomic status as preventive factors for mild cognitive impairment and Alzheimer’s disease. Psychiatry Res. 2012;196(1):90–5. [DOI] [PubMed] [Google Scholar]
- 41.Wang T, Mao L, Wang J, Li P, Liu X, Wu W. Influencing factors and exercise intervention of cognitive impairment in elderly patients with chronic obstructive pulmonary disease. Clin Interv Aging. 2020;15:557–66. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Baird C, Lovell J, Johnson M, Shiell K, Ibrahim JE. The impact of cognitive impairment on self-management in chronic obstructive pulmonary disease: A systematic review. Respir Med. 2017;129:130–9. [DOI] [PubMed] [Google Scholar]
- 43.Yadav UN, Lloyd J, Hosseinzadeh H, Baral KP, Harris MF. Do chronic obstructive pulmonary diseases (COPD) Self-Management interventions consider health literacy and patient activation?? A systematic review. J Clin Med. 2020;9(3):646. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Jassem E, Kozielski J, Górecka D, Krakowiak P, Krajnik M, Słomiński JM. Integrated care for patients with advanced chronic obstructive pulmonary disease: a new approach to organization. Pol Arch Med Wewn. 2010;120(10):423–8. [PubMed] [Google Scholar]
- 45.Elbeddini A, Tayefehchamani Y. Amid COVID-19 pandemic: challenges with access to care for COPD patients. Res Social Adm Pharm. 2021;17(1):1934–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Wei L, Champman S, Li X, Li X, Li S, Chen R, et al. Beliefs about medicines and non-adherence in patients with stroke, diabetes mellitus and rheumatoid arthritis: a cross-sectional study in China. BMJ Open. 2017;7(10):e017293. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Data is provided within the manuscript or supplementary information files.