Do

ChHa Cb Dc Dd Ne G

International Journal of Nursing Studies 50 (2013) 90–99

A

Art

Re

Re

Ac

Ke

He

Nu

Sca

Ty

*

00

htt

evelopment and preliminary testing of a brief screening measuref healthy lifestyle for diabetes patients

ia-Pei Chen a, Yun-Shing Peng b, Hsu-Huei Weng c, Jun-Yu Fan d, Su-Er Guo d,siao-Yun Yen a, Yun-Feng Tseng a, Mei-Yen Chen e,*

hang Gung Memorial Hospital, Chiayi, Taiwan, ROC

ivision of Endocrinology and Metabolism, Chang Gung Memorial Hospital, Chiayi, Taiwan, ROC

epartment of Diagnostic Radiology, Chang Gung Memorial Hospital, Taiwan, ROC

ursing Department, Chang Gung University of Science and Technology, Taiwan, ROC

raduate Institute of Nursing, Chang Gung University of Science and Technology, Taiwan, ROC

R T I C L E I N F O

icle history:

ceived 17 March 2012

ceived in revised form 2 September 2012

cepted 3 September 2012

ywords:

alth promotion

rsing

le development

pe 2 diabetes

A B S T R A C T

Objectives: Improving diabetes control is a critical issue world-wide, because mortality

rates and costs are increasing dramatically. Maintaining a healthy lifestyle is positively

associated with diabetes control. However, the availability of practical screening tools

suitable for determining and modifying healthy or unhealthy lifestyle choices is limited.

The main aim of this study was to determine the appropriateness of the type 2 diabetes

and health promotion scale (T2DHPS) for use in patients with type 2 diabetes.

Methods: This study examined the Cronbach’s alpha, content validity, construct and

concurrent validity of the Chinese language version of the T2DHPS for assessing lifestyle

and disease control among patients with type 2 diabetes. The dimensions of the T2DHPS

were generated from the Chinese version of adult health promotion and interviews with

experts, and were corroborated by the literature. A total of 323 patients previously

diagnosed with type 2 diabetes were recruited. A cross-sectional, descriptive design

questionnaire was developed and tested at diabetes outpatient departments in three

teaching hospitals between August 2010 and June 2011. Construct validity was

established using factor analysis. The total and subscale scores of the T2DHPS were

correlated with biomarkers of diabetes control for concurrent validity.

Results: Kaiser–Meyer–Olkin (KMO) and Bartlett’s sphericity tests showed that the sample

met the criteria required for factor analysis. A 28-item Likert-type scale of the T2DHPS was

established, and explained 56.7% of the total variance. The simplified version of the

T2DHPS was made up of six dimensions of behavior: physical activity, risk reduction,

stress-management, enjoy life, health responsibility and a healthy diet. The reliability

coefficient for the total scale was 0.89, and alpha coefficients for the subscales ranged from

0.63 to 0.86. Concurrent validity indicated that the T2DHPS is significantly positively

associated with diabetes control.

Conclusions: T2DHPS was shown to be a reliable and valid tool for assessing patients with

type 2 diabetes, and can possibly predict diabetic control. This scale appears to be a useful

screening tool for type 2 diabetic people in primary health care settings, to promote health

status through modification of an unhealthy lifestyle.

� 2012 Elsevier Ltd. All rights reserved.

Corresponding author at: No. 2, Chia-pu Rd., West Sec., Putz City, Chiayi County 61363, Taiwan, ROC. Tel.: +886 5 3628800x2201; fax: +886 5 3628866.

E-mail address: meiyen@gw.cgust.edu.tw (M.-Y. Chen).

Contents lists available at SciVerse ScienceDirect

International Journal of Nursing Studies

journal homepage: www.elsevier.com/ijns

20-7489/$ – see front matter � 2012 Elsevier Ltd. All rights reserved.

p://dx.doi.org/10.1016/j.ijnurstu.2012.09.001

�

�

�

�

�

�

1

(TmOwhthtothdthHaospc1mtha

pc2bh[A1imainfub2in

C.-P. Chen et al. / International Journal of Nursing Studies 50 (2013) 90–99 91

What is already known about the topic?

Many studies have indicated that adopting a healthylifestyle is beneficial for type 2 diabetes patients.

Previous studies have focused on numerous diabetesmellitus patients and their lifestyle changes worldwide.

Many studies have focused on assessing the psychosocialand separate health-related behaviors among type 2diabetes patients. However, this has been limited by alack of valid screening tools to detect unhealthy lifestylesamong diabetes sufferers.

What this paper adds

This study provides a quick, easy, practical andcomprehensive screening tool of health promotionoriented behaviors for type 2 diabetes patients.

The results show that the T2DHPS is significantlyassociated with diabetes control.

This study suggests that the T2DHPS can be used for theearly detection and modification of unhealthy lifestylesin type 2 diabetes sufferers.

. Introduction and background

The increasing prevalence of type 2 diabetes mellitus2DM) and an associated increase in mortality poses aajor public health challenge worldwide [World Healthrganization (WHO), 2012]. Some 346 million peopleorldwide have diabetes, and an estimated 3.4 million

ave died from the consequences. Diabetes can damagee heart, blood vessels, eyes, kidneys, and nerves and lead

foot amputation and associated deaths. Furthermore,e WHO projects that diabetes-associated mortality will

ouble between 2005 and 2030 (WHO, 2012). According toe Taiwan National Health Statistics [Department ofealth (DOH), Taiwan, 2012], the prevalence of T2DM atge 40–59 was 11.37%, increasing to 21.82% for 60-year-lds. According to the statistics of the internationaltatistical classification of diseases and related healthroblems, 10th revision (ICD-10), T2DM is the 4th mostommon cause of death, and accounted for 26.9 deaths per00,000 in 2011 (DOH, Taiwan, 2012). The standardizedortality rate from T2DM in Taiwan is higher than that ine United States, the United Kingdom, Singapore, Korea

nd Japan (WHO, 2012).There is plenty of evidence that a healthy lifestyle is

ositively associated with health status and glycemicontrol of type 2 diabetes all over the world (Greaves et al.,011). Studies have demonstrated that ‘healthy eating’ehaviors could lead to significantly lower glycosylatedemoglobin (HbA1c) levels among diabetic patientsmerican Diabetes Association (ADA), 2011]. A weekly

50-minute moderate-intensity aerobic program canprove glycemic measures, particularly among middle-

ged and older adults with type 2 diabetes. The ‘moderate-tensity aerobic program’ includes cardiopulmonarynction-improving activities such as brisk walking,

iking, badminton, tai chi, and aerobics (Chiu and Wray,010; Greaves et al., 2011; Sakane et al., 2011), and

groups (Elsawy and Higgins, 2010). In addition, maintain-ing a normal body weight and avoiding tobacco use canprevent or delay the onset of type 2 diabetes and itscomplications (WHO, 2012).

The health status of diabetic patients is mainly assessedby measuring fasting blood glucose (FBG), HbA1c, bloodpressure, and low density lipoprotein-cholesterol (LDL-C)(ADA, 2011; DOH, 2012). The ideal levels for positivediabetes control should be: HbA1C < 7.0%, fasting bloodglucose (FG) < 130 mg/dl, LDL < 100 mg/dl and bloodpressure < 130/85 mmHg. Reduced incidences of dia-betes-related complications have been reported whenblood glucose, blood pressure and blood lipid are wellcontrolled [Bureau of Health Promotion (BHP), 2012].

The American Association of Diabetes Educators (AADE,2008) has suggested seven self-care behaviors as a newparadigm for diabetes education, including healthy eating,physical activity, monitoring, taking medications, problemsolving, stress management and risk reduction. Althoughfavorable health behaviors for patients with T2DM havebeen suggested in many studies, the broad spectrum of thisadvice was derived from various perspectives on diet,exercise, oral hygiene, foot care, medication, risk taking,social support, emotional status, empowerment, and self-care. As a consequence, the availability of a simple andshort screening tool for assessing the association of healthbehaviors and related health status in routine practice islimited.

To date, more than 26 instruments have been devel-oped to assess and evaluate lifestyles and adaptation fordiabetes people around the world. These have included:

(1) Knowledge, attitude, empowerment and self-efficacyassessment related to diabetes. The concepts of theabove scales include e.g. the mechanism and physiol-ogy of diabetes, insulin reaction, how to deal with acuteand chronic complications, motivation for diabetesmanagement, emotional adjustment, assessing psy-chological distress, and confidence in managingdiabetes (Daly et al., 2009; Dunn et al., 1986; Fitzgeraldet al., 1998; Polonsky et al., 2005; Rapley et al., 2003;Shiu et al., 2003).

(2) Diabetes care profile, e.g. medication use, mealplanning, exercise, blood glucose monitoring, dietrestriction, compliance with medication, foot careand smoking cessation (Anderson et al., 2003; Boyerand Earp, 1997; Fitzgerald et al., 1996; Meadows et al.,2000; Toobert et al., 2000; Wong et al., 2009). (3)Diabetes and quality of life, for instance diabetescontrol, satisfaction with treatment, social life andsexual function (Bradley et al., 1999; Hwang et al.,2009; Redekop et al., 2002).

Summarizing the above references, most of the assess-ment scales consist of more than 50 items (range from 8 to234 items) and many of them are focused on disease-oriented rather than health promotion behaviors. Further-more, many important concepts and behaviors are spreadover different assessment scales. For example, medication,diet, exercise, risk avoidance, regular check-up, emotional

nd psychosocial distress, oral hygiene and foot care are

volves repeated exercise mediated by the large muscle a

imetasmcaastafobuemthpspr

2.

2.1

evhesupadicethcrphfupaaninbe

nemwreodThcoaspr

2.2

sythprsetodi–

anCoinThinte

du

C.-P. Chen et al. / International Journal of Nursing Studies 50 (2013) 90–9992

portant issues for glycemic control (AADE, 2008; Daly al., 2009; Wang et al., 2008), but the availability of ansessment scale that puts them together as a screeningeasure is limited. In consequence, the primary healthre providers in Taiwan tend to miss the comprehensivesessment, and lack ability to provide individuallyilored care for their diabetes clients. For instance, somecus on taking medication, diet and testing blood glucose,t lack the assessment of exercise, stress management,otional aspect and oral hygiene or foot care. Therefore,

e purpose of this study was to establish and assess theychometric properties of a type 2 diabetes and healthomotion scale (T2DHPS) among patients with diabetes.

Methodology

. Design, sample and setting

This study was part of a larger clinical study gatheringidence of health-promoting behavior associated withalth status in type 2 diabetes patients. A cross-sectionalrvey and descriptive design were used in this study. Therticipants were selected by convenience sampling atabetes outpatient departments from three medicalnters in Northern, Central and Southern Taiwan duringe period from August 2010 to June 2011. The inclusioniteria were: (1) patients diagnosed with T2DM byysicians, (2) adults aged 20 years and over, (3) patients

lly independent in managing their daily lives, (4)tients able to complete the questionnaires in Mandarind Hokkien dialects either by self-administration or atterview, and (5) who agreed to sign an informed consentfore being enrolled into the study.The determination of sample size was based on the

ed to perform factor analysis, which requires a mini-um of 300 samples (Comrey and Lee, 1992). The T2DHPSas evaluated for reliability and validity. Split-halfliability was computed by correlating the scores of thed half with those of the even half of the total T2DHPS.e internal consistency was assessed by Cronbach’s aefficient. Construct validity and concurrent validity weresessed by correlation analysis and factor analysis usingincipal component analysis (Hair et al., 1998).

. Procedure and ethical considerations

There were four stages in the set-up of this study: (1)stematic literature reviews to draw up the first version ofe questionnaire. Type 2 diabetes, healthy lifestyle, healthomotion and scale were used as the major keywords toarch PubMed, Ovid and the Chinese database from 1980

2011. (2) Eleven experts – metabolic physicians,abetes educators, and nurses who teach diabetes carewere invited to audit the contents. (3) Construct validityd reliability were established through factor analysis. (4)ncurrent validity was determined by comparing the

dividual results with each participant’s glycemic control.e HbA1C and FBG data were collected on the day of

terview and the results of the latest blood laboratorysts within three months.

Before this study was started, the interviewing proce-

approved by the researchers’ Institutional Review Board(No. 98-1452B). The study objectives were explained topotential participants at interview by one of four seniorcertified diabetes educators. Signed informed consents wereobtained from those who agreed to participate in the study.

2.3. Data analysis

Descriptive and inferential statistics were processedusing SPSS 17.0 software. All tests were 2-sided and p-values < 0.05 were considered statistically significant. Theprincipal component factor analysis was performed usingvarimax rotation with Kaiser normalization. Pearson’sproduct moment correlation coefficient was used toevaluate concurrent validity.

2.4. Item development

Primary identification of potential items to be incorpo-rated into the T2DHPS was accomplished using a review ofrelated literature (AADE, 2008; ADA, 2011; Chen and Chou,1999; CDC, 2011). The initial form of the T2DHPScomprised 62 items relating to adults’ health-promotingbehaviors and diabetes care, with ten categories:

(1) Healthy diet, which included eight items, e.g., I eat abalanced diet every day; I have five groups in eachmeal per day; I have three portions of vegetables perday; I avoid eating foods high in fat; I control my dietwhen eating outside or on special days.

(2) Physical activity, which included five items, e.g.,regular exercise at least 30 minutes per day, or150 minutes per week; I still do exercise, even thoughI work a lot.

(3) Social support, which included five items, e.g., I doexercise with my family/or friends; I have maintainedactivities with my friends, even after being diagnosedwith diabetes.

(4) Health responsibility, which included nine items, e.g.,I have my eyes examined periodically; I visit a doctorperiodically; I eat something before doing exercise; Ihave my blood lipids examined periodically; I havemy blood glucose examined periodically; I have ablood glucose machine at home.

(5) Stress management, which included four items, e.g., Itry to relax when I am in a bad mood (listen to musicor sing a song); when I feel under pressure I try tounderstand the reasons and face them; I sleep at least6–8 hours per day.

(6) Enjoy life, which included five items, e.g., I believe thatmy life has purpose; I am content with myself; I havecontinued to work since being diagnosed withdiabetes.

(7) Medication, which included five items, e.g., I followthe physician’s prescriptions; I do not change themedication myself; I do not buy drugs from thedrugstore myself; I still maintain my use of medica-tion.

(8) Risk-reducing behaviors, which included eight items,e.g., I know how to deal with hyperglycemia and

res, privacy protection and ethical considerations were

hypoglycemia; I check my toes and feet to see if there

(1

ovtinaT1eavo((eitbpa6

oppdrc

3

3

lerth(Sw(S5le

3

sinarrccin

C.-P. Chen et al. / International Journal of Nursing Studies 50 (2013) 90–99 93

are any wounds; I try not to get exposed to second-hand smoke.

(9) Oral hygiene, which included five items, e.g., I brushmy teeth after meals; I brush my teeth at least twiceper day; and

0) Foot care, which included eight items, e.g., I wearslippers indoors and shoes outdoors; I do foot andankle exercises daily; I try not to cross my feet.

The questionnaire used a 5-point response format tobtain data regarding the frequency of reported beha-iors (never, rarely, sometimes, usually, always), withhe rating score ranging from 1 to 5. Eleven experts were

volved in the professional critiques. The experts weresked to rate each item of the Chinese version of the2DHPS on the basis of relevance and appropriateness as (not relevant), 2 (relevant) or 3 (very relevant). Thexperts were also asked to feed back editorial changesnd add comments or items to the blank sheet. A contentalidity index (CVI) was computed using the proportionf experts who were in agreement about item relevancePolit and Beck, 2008). The average content validity indexCVI) was 0.97. On the basis of feedback from the contentxperts, editorial changes were made and seven newems were inserted into the T2DHPS. For instance, Irought sugar with me; I will talk to health careersonnel if I notice hypoglycemia symptoms; I exerciset home on a bad day. The final questionnaire contained9 items.

Furthermore, as a pilot study to establish the suitabilityf the questionnaire and the time spent on it by thearticipants, this revised T2DHPS was administered to 10atients with type 2 diabetes at the diabetes outpatientepartment. Most of them took about 20–30 minutes toeview the questionnaire and agreed that most of theontent items matched their lifestyle and situations.

. Results

.1. Demographic characteristics of the sample

Four cases failed to complete the questionnaires,aving 323 subjects with valid data. Almost all of the

espondents (N = 306, 94.7%) were Taiwanese. The ages ofe subjects ranged from 21 to 87 years with a mean of 59.5D = 11.3) and a median of 60. Of the 323 patients, 65.9%ere male. The mean duration of diabetes was 8.6 yearsD = 7.6). Most participants were overweight (67%), and

0.4% had been educated only to primary school level orss (receiving less than six years).

.2. Item analysis

Before the item analysis was used, 11 items wereeldom reported as habits, with few of the subjectsdicating that they had behaviors in these items such

s chewing betel nuts, using tobacco, drinking alcohol, oreceiving flu vaccinations each year. Therefore, 58 itemsemained. Later, serial calculations of corrected item-totalorrelation coefficients identified 21 items that did notontribute to the internal consistency of the overallstrument or the subscales. These 21 items were

eliminated from the 58-item questionnaire. Many of theitems pertaining to medication use, oral hygiene and risk-reducing dimensions were deleted, for example: I followthe physician’s prescription, I do not change the medica-tion myself, I do not buy drugs from the drugstore myself, Iknow how to deal with hyperglycemia, I know how to dealwith hypoglycemia, I don’t use scissors to cut my nails, Idon’t use hot water or a heater to warm my body on a coldday, I don’t smoke cigarettes, I avoid exposure to second-hand smoke, I choose foods without preservatives.

Of the remaining 37 items, 31 had item-total correla-tions of 0.35 or higher. The interim correlation matrix wasexamined to identify items that were possibly redundantand therefore could be eliminated to shorten the ques-tionnaire. Six items with correlations above 0.70 werefound and deleted to prevent multicollinearity of themultivariate analysis (Hair et al., 1998); hence, 31 itemswere retained for the next stage of the analysis.

3.3. Factor analysis

Before the exploratory analysis, Kaiser–Meyer–Olkin(KMO) and Bartlett’s sphericity tests were used to measurethe sampling adequacy. The results showed that the KMOvalue was 0.897, and the significance of Bartlett’ssphericity was <0.001 (x2 = 3486.52, p < 0.001), indicatingthat the sample met the criteria for factor analysis (Hairet al., 1998). Principal component factor analysis wasperformed using varimax rotation with Kaiser normal-ization. Factor analysis yielded a 9-factor solution with anexplained variance of 63.71%, which had eigenvaluesgreater than 1.00. To define the composition of the ninefactors more efficiently, items that did not load strongly orcleanly on a single factor were eliminated. According to thescree plot (Fig. 1), the slope of the curve became emergentat the sixth point, and factors 7 through 9 only contributed7% of the accumulated variance. Three items wereeliminated because they did not load strongly on a singlefactor, and a six-factor instrument was decided upon. Theremaining 28 items were entered into the factor analysis.All items loaded on expected factors, and the variance was0.50 for most items, with only two items below 0.4 (Table1). Factor analysis of the resulting 28-item instrumentyielded a six-factor solution with an explained variance of56.66%, as shown in Table 2.

Factor 1 is physical activity behavior. This was thestrongest factor, showing the greatest percentage variancein the T2DHPS. Item loading on this factor includes sevenitems: I spare time for exercise even if I am very busy, I stillcontinue doing exercise when exercising does not seem tohelp my weight, I do moderate-intensity exercise morethan 150 minutes every week, I still do exercise even I amoccupied with a lot of work every day, I do exercise indoorson bad-weather days, I eat something first before doingexercise, and I do exercise with my family or friends. Factor2 is risk reducing behavior and includes the followingseven items: I check my toes and feet to see if there are anywounds, I read diabetes information from newspaper ormagazines, I brush my teeth after meals, I check if there areany little pieces on my feet before putting on shoes, I readfood labels when I shop (e.g. ingredients, calories), I wear

sliex

item

Ta

Fac

V

P1

2

3

4

5

6

7

R8

9

1

1

1

1

1

S1

1

1

1

1

E2

2

2

H2

2

2

H2

2

2

C.-P. Chen et al. / International Journal of Nursing Studies 50 (2013) 90–9994

ppers indoors and shoes outdoors, I do foot and ankleercises every day.Factor 3, stress management behaviors, included fivems: I try to relax when I am in a bad mood, e.g., listen to

usic or sing a song, when I feel under pressure I try to

understand the reasons and face them, I have continued towork after being diagnosed with diabetes, I have main-tained activities with friends after being diagnosed withdiabetes, I arrange my daily life well so that I am not in ahurry. Factor 4 is enjoyment of life and includes the

Fig. 1. The scree plot.

ble 1

tor loadings and factor structure of T2DHPS.

ariables 1 2 3 4 5 6

hysical activity. I spare time for exercise even if I am very busy .849

. I still continue doing exercise when exercising does not seem to help my weight .796

. I do moderate-intensity exercise more than 150 minutes every week .791

. I still do exercise even I am occupied with a lot of work every day .787

. I do exercise indoor on bad-weather days .694

. I eat something first before doing exercise .511

. I do exercise with my family or friends .466

isk reduction. I check my toes and feet to see if any wounds .678

. I read diabetes information from newspaper or magazines .630

0. I brush my teeth after meals .616

1. I check if there are any little pieces on my feet before putting on shoes .605

2. I read food labels when I shop (e.g. ingredients, calories) .565

3. I wear slippers indoors and shoes outdoors .553

4. I do foot and ankle exercises every day .349

tress management5. I try to relax when I am in a bad mood, e.g., listen to music or sing a song .771

6. I try to understand the reasons and face them when I feel under pressure .751

7. I have continued to work after being diagnosed with diabetes .614

8. I have maintained activities with friends after being diagnosed with diabetes .492

9. I arrange my daily life well, so that I am not in a hurry .392

njoy life0. I believe that my life has purpose .830

1. I am content with myself generally speaking .796

2. I have paid attention to my health since being diagnosed with diabetes .694

ealth responsibility3. I have my eyes examined periodically .671

4. I visit a doctor periodically .630

5. I have my blood lipids measured periodically .629

ealthy diet6. I control my diet while eating outside or on special days .772

7. I avoid eating foods high in fat (e.g., fried or fat meat) .722

8. I have a balanced diet every day .509

foaad

indpino(e

3

inthcA0Tr

T

In

T

C

T

E

o

C.-P. Chen et al. / International Journal of Nursing Studies 50 (2013) 90–99 95

llowing three items: I believe that my life has purpose, Im content with myself generally speaking, I have paidttention to my health since being diagnosed withiabetes.

Factor 5 contains health responsibility behaviors andcludes: I have my eyes examined periodically, I visit a

octor periodically, and I have my blood lipids measurederiodically. Finally, factor 6 is healthy diet, whichcorporates three items: I control my diet while eating

utside or on special days, I avoid eating foods high in fat.g., fried or fat meat), I have a balanced diet every day.

.4. Reliability

Cronbach’s alpha was calculated as a measure ofternal consistency for the final 28-item instrument;e total questionnaire was found to have high internal

onsistency, with an alpha coefficient of 0.89 (Table 3).lpha coefficients for the six subscales ranged from 0.63 to.86. The split-half reliability coefficients of the 28-2DHPS were 0.85 and 0.81 for part 1 and part 2espectively.

3.5. Scoring

The questionnaire used a Likert-type scale with therating score ranging from 1 to 5. When the instrument wasused as a whole, the possible range of scores was from 28 to140. Evaluation of the 323 individuals responding to theT2DHPS showed that the total mean score was 96.49(SD = 18.50), with a range from 19 to 139 and a median of98. Histograms of the distributions of scores on the totalinstrument and the six subscales were fairly symmetrical(skewness = �0.596, kurtosis = 0.305) within the range ofscores used.

3.6. Concurrent validity

Table 4 shows that the T2DHPS and six sub-scales wereall negatively associated with FBG and HbA1C. The totalscore of the T2DHPS (p < 0.05), physical activity (p < 0.05),and risk reduction (p < 0.05) were statistically significantlyassociated with FBG. Further, except for the sub-scale ofrisk reduction, the total score of the T2DHPS (p < 0.001),physical activity (p < 0.001), enjoying life (p < 0.01), stressmanagement (p < 0.01), health responsibility (p < 0.001),and healthy eating behaviors (p < 0.05) were statisticallysignificantly associated with A1C.

4. Discussion

The limitations of this study include: (1) participantswere invited by nonrandom sampling and most of themapparently had lower social economic status, such asreceiving less than six years of education. This might limitthe generalizability of the findings. (2) Selection and recallbias need to be considered because all the participants hadsuffered various durations of diabetes, times and levels ofhealth condition. (3) Test–retest reliability is an importantproperty for development of a scale but was not measuredin this study (Polit and Beck, 2008). Further researchshould also consider designing a longitudinal study andrandom sampling.

Despite the limitations, the study showed that theT2DHPS provides a quick, practical guide and comprehen-sive screening tool for health promotion-oriented beha-viors for type 2 diabetes patients. In addition, it can be usedfor the early detection and modification of unhealthylifestyles in type 2 diabetes. The validity of the T2DHPS wasestablished through content validation by a consensus ofdiabetic experts, construct validation using the explora-tory factor analysis, and correlation with glycemicbiomarkers as concurrent validation. The high Cronbach’salpha indicated that the T2DHPS was adequate in terms ofinternal consistency.

At the beginning of the questionnaire development, thepilot form of the original scale incorporated healthpromotion and diabetes care concepts with the 69 items,but more than half of these items were deleted in the finalversion. The items were deleted on the basis of their lowercontribution to the total correlation coefficients. Surpris-ingly, some important items were deleted, including: Icomply with medication prescribed by a physician; I do notchange the medication myself, even if my blood glucose is

able 3

ternal consistency of the T2DHPS and its subscale.

Subscale Items Item-subscale r Cronbach’s a

Physical activity 7 0.504–0.868 0.86

Risk reduction 7 0.518–0.719 0.73

Stress management 5 0.643–0.837 0.79

Enjoy life 3 0.383–0.883 0.82

Health responsibility 3 0.609–0.842 0.63

Healthy diet 3 0.738–0.838 0.68

Total scale 28 0.89

able 4

oncurrent validity of T2DHPS and diabetes control.

Variables FBGa A1C

Physical activity �.142* �.215***

Risk reduction �.148* �.109

Enjoy life �.050 �.166**

Stress management �.043 �.189**

Health responsibility �.036 �.262***

Healthy eating �.085 �.122*

Total score �.150* �.249***

a FBG: fasting blood glucose.

* p < .05, Pearson correlation, two-tailed.

** p < .01, Pearson correlation, two-tailed.

*** p < .001, Pearson correlation, two-tailed.

able 2

igen values, cumulative percentage of variance explained by six factors

n the T2DHPS.

Factor Factor label Eigen

value

Variance

explained

Cumulative

percentage

1 Physical activity 4.05 14.45 14.45

2 Risk reduction 2.81 10.04 24.49

3 Stress management 2.72 9.72 34.20

4 Enjoy life 2.46 8.77 42.98

5 Health responsibility 1.99 7.11 50.08

6 Healthy diet 1.84 6.58 56.66

stamstu(2bebladasdidihehedopepacoe.ge.gtorefualshesc

wprboretoglglplthprofthpa

heam

C.-P. Chen et al. / International Journal of Nursing Studies 50 (2013) 90–9996

ble; I still maintain use of the medication; and I checky blood glucose. This finding is inconsistent with otherdies (Daly et al., 2009; Toobert et al., 2000). Daly et al.

009) and Toobert et al. (2000) indicated that self-carehaviors consist in adherence to medication use andood glucose monitoring. In their study, self-reportedherence with medication and blood glucose testing wassociated with lower HbA1C. A possible reason for thescrepancy is that these behaviors were included in themension of health responsibility because in Taiwan’salth care system, 99% people are covered by the nationalalth insurance (DOH, 2012). Therefore, when ‘‘I visit actor periodically and I have my blood lipids measuredriodically’’, physicians will also regularly check thetient’s blood glucose and prescribe medication. Thensequence is eliminated by high inter-item correlation,. r > 0.7, or low contribution to item-total correlation,. r < 0.35. However, this phenomenon necessarily needs

be explored in a further study. On the other hand,ceiving flu vaccinations, checking blood pressure, renalnction, waist circumference, and dental check-up wereo omitted. This trimming was necessary as, in standardalth care settings, it is not acceptable for assessmentales to use too many items to assess a client’s lifestyle.

The concurrent validity in a study is demonstratedhen a test correlates well with a measure that haseviously been validated (Hair et al., 1998); thus, usingth HbA1C and FBG to compare with the questionnairesults is reasonable for assessing the validity of this newol. According to the recommendations for assessment ofycemic control, monitoring both the fasting blooducose and HbA1c are priorities for the managementan of diabetes. Additionally, HbA1c is thought to reflecte average glycemia over several months, and is a strongedictor of diabetes complications (ADA, 2011). The result

the bivariate correlation indicated that the total score ofe T2DHPS and the sub-scale significantly reflected therticipant’s diabetes control.There are many studies that have shown that adopting a

althy lifestyle is positively associated with health statusong type 2 diabetes patients (ADA, 2011; Chiu and

Wray, 2010). Hence, lifestyle intervention studies fordiabetes care are being conducted in many countries(Linmans et al., 2011; Sakane et al., 2011). According tosystematic reviews (Greaves et al., 2011), most of thelifestyle intervention components comprised diet andphysical activity, with measurement outcomes beingweight loss and physical activity behavior changes.Recently, some of these studies have shown that lifestyleintervention programs may only have a small effect inreducing FBG and A1C in diabetes clients (Linmans et al.,2011). Therefore, searching for an effective, simple,practical guide for lifestyle assessment and interventionis still a challenge for our primary health care providers.The findings indicate that this T2DHPS might provide aclinical contribution with another perspective on healthpromotion.

5. Conclusion

This study showed that the T2DHPS is a small scale (seeAppendix A), valid and reliable instrument for evaluatingthe extent of healthy lifestyles for type 2 diabetes adults.The strength of this study is its focus on positive elementsfor health promotion. Specific interventions can bedeveloped further to promote healthy behaviors adoptedby diabetes people and increase their positivity and self-worth. This combination makes it a practical guide foridentifying and modifying unhealthy behavior for diabetespatients.

Conflict of interest. The author(s) declare that they have no

competing interests.

Funding. The study was supported by a grant from the foun-

dation of Chang Gung Memorial Hospital (CMRPG680501).

Ethical approval. This study was approved by the institutional

review board of the ethical committee of Chang Gung Mem-

orial Hospital (IRB 98-1452B).

Appendix A

(Type 2 Diabetes and Health Promotion Scale)

%)Directions: This questionnaire contains

statements regarding your present way of life or personal habits. Please respond to each item as accuratelyas possible, and try not to skip any ite m. Please fill in the av ailable blank that you did during this ye ar).

10%____30%____ 50%_ __70%__ _90%

(Questions)

Never Sometimes Half Often Always

1. I spare time for exercise even if I am very busy.

2. I still continue doing exercise when exercising does not seem to

help my we ight.

3. I do moderate-intensity exercise more than 150 minutes every week .

4. I still do exercise even I am occupied with a lot of work every day.

5. I do exercise indoor on bad-weather days.

6. I eat something first before doing exercise.

7. I do exercise with my family or friends.

8. I check my toes and feet to see if any wounds.

9. I read diabetes information from newspaper or magazines.

10. I brush my teeth after meals.

11. I check if there are any little pieces on my feet before putting on

shoes.

12. I read food labels when I shop (e.g., ingredients, calories).

13. I wear slippers indoors and shoes outdoors.

C.-P. Chen et al. / International Journal of Nursing Studies 50 (2013) 90–99 97

Re

Am

Am

An

1

1

1

1

1

1

t

2

2

2

2

2

2

2

2

2

C.-P. Chen et al. / International Journal of Nursing Studies 50 (2013) 90–9998

ferences

erican Association of Diabetes Educators (AADE), 2008. AADE7TM Self-Care Behaviors. The Diabetes Educator 34 (3), 445–449.erican Diabetes Association, ADA, 2011. Standards of medical care indiabetes-2011. Diabetes Care 34 (1), S4–S61.

derson, R.M., Fitzgerald, J.T., Gruppen, L.D., Funnell, M.M., Oh, M.S.,2003. The diabetes empowerment scale-short form (DES-SF). Dia-

Bureau of Health Promotion (BHP), 2012. Adult and Elderly Health. Depart-ment of Health, Taiwan. , http://www.bhp.doh.gov.tw/BHPnet/Portal/Them_Show.aspx? Subj ect=200712250011&Clas s=2&No=201110260001.

Bradley, C., Todd, C., Gorton, T., Symonds, E., Martin, A., Plowright, R.,1999. The development of an individualized questionnaire measureof perceived impact of diabetes on quality of life: the ADDQoL-Qualityof life research: an international journal of quality of life aspects of

4. I do foot and ankle exercises every day.

5. I try to relax when I am in a bad mood, e.g., listen to music or sing

a song.

6. I try to understand the reasons and face them when I feel under

pressure.

7. I have continued to work after being diagnosed with diabetes.

8. I have maintained activities with friends after being diagnosed with

diabetes.

9. I arrange my daily life well (e.g. medication, exercise and work), so

hat I am not in a hurry.

0. I believe that my life has purpose.

1. I am content with myself generally speaking.

2. I have paid attention to my health since being diagnosed with

diabetes.

3. I have my eyes examined periodically.

4. I visit a doctor periodically.

5. I have my blood lipids measured periodically.

6. I control my diet while eating outside or on special days.

7. I avoid eating foods high in fat (e.g., fried or fat meat).

8. I have a balanced diet every day.

treatment. Care and Rehabilitation 8 (1–2), 79–91.

betes Care 26 (5), 1641–1642.

B

C

C

C

C

D

D

D

E

F

F

G

H

H

C.-P. Chen et al. / International Journal of Nursing Studies 50 (2013) 90–99 99

oyer, J.G., Earp, J.A., 1997. The development of an instrument for asses-sing the quality of life of people with diabetes-39. Medical Care 35 (5),440–453.

enter for Disease Control and Prevention (CDC), 2011. Chronic DiseasePrevention and Health Promotion: Preventing Cavities, Gum Disease,and Tooth Loss. , http://www.cdc.gov/nccdphp/publications/aag/doh.htm.

hen, M.Y., Chou, C.C., 1999. The Chinese version of adult’s healthpromotion scale development. Chang Gung Nursing 8 (1), 14–23.

hiu, C.J., Wray, L.A., 2010. Factors predicting glycemic control in middle-aged and older adults with type 2 diabetes. Preventive ChronicDisease 7, A08.

omrey, A.L., Lee, H.B., 1992. A First Course in Factor Analysis. Erlbaum,Hillsdale, NJ.

aly, J.M., Hartz, A.J., Xu, Y., Levy, B.T., James, P.A., Merchant, M., Garrett,R.E., 2009. An assessment of attitudes and behaviors of patients withtype 2 diabetes and outcomes. Journal of the American Board ofFamily Medicine 22 (3), 280–290.

epartment of Health, Executive Yuan, Taiwan, ROC, 2012. Annual Reportfor Health Statistics of National Health Insurance-2011. , http://www.doh.gov.tw/CHT2006/DM/DM2_2_p02.aspx?class_no=440&now_fod_list_no=11587&level_no=3&doc_no=79126.

unn, S.M., Smartt, H.H., Beeney, L.J., Turtle, J.R., 1986. Measurement ofemotional adjustment in diabetic patients: validity and reliability ofATT-39. Diabetes Care 9 (5), 480–489.

lsawy, B., Higgins, K.E., 2010. Physical activity guidelines for older adults.American Family Physician 81, 55–59.

itzgerald, J.T., Davis, W.K., Connell, C.M., Hess, G.E., Funnell, M.M., Hiss,R.G., 1996. Development and validation of the diabetes care profile.Evaluation & The Health Professions 19 (2), 208–230.

itzgerald, J.T., Funnell, M.M., Hess, G.E., Barr, P.A., Anderson, R.M., Hiss,R.G., 1998. The reliability and validity of a brief diabetes knowledgetest. Diabetes Care 21 (5), 706–710.

reaves, C.J., Sheppard, K.E., Abraham, C., Hardeman, W., Roden, M.,Evans, P.H., Schwarz, P., 2011. Systematic review of interventioncomponents associated with increased effectiveness in dietary andphysical activity interventions. BMC Public Health 11, 119.

air, J.F., Anderson, R.E., Tatham, R.L., Black, W.C., 1998. Multivariate DataAnalysis, 5th ed. Prentice-Hall, Englewood Cliffs, NJ.

wang, C.C., Huang, I.C., Wu, M.Y., Chen, T.C., Wu, J.J., Lin, W.D., 2009.Development and validation of a Chinese version of the diabetes-39 to

measure diabetes quality of life in Taiwan. Taiwan Journal of PublicHealth 28 (3), 218–231.

Linmans, J.J., Spigt, M.G., Deneer, L., Lucas, A.E., de Bakker, M., Gidding,L.G., Linssen, R., Knottnerus, J., 2011. Effect of lifestyle intervention forpeople with diabetes or prediabetes in real-world primary care:propensity score analysis. BMC Family Practice 12, 95.

Meadows, K.A., Abrams, C., Sandbaek, A., 2000. Adaptation of the diabeteshealth profile (DHP-1) for use with patients with type 2 diabetesmellitus: psychometric evaluation and cross-cultural comparison.Journal of the British Diabetic Association 17 (8), 572–580.

Polit, D.F., Beck, C.T., 2008. Nursing Research: Principles and Methods, 8thed. Lippincott Williams and Wilkins, Philadelphia.

Polonsky, W.H., Fisher, L., Earles, J., Dudl, R.J., Lees, J., Mullan, J., 2005.Assessing psychosocial distress in diabetes: development of thediabetes distress scale. Diabetes Care 28 (3), 626–631.

Rapley, P., Passmore, A., Phillips, M., 2003. Review of the psychometricproperties of the diabetes self-efficacy scale: Australian longitudinalstudy. Nursing & Health Sciences 5 (4), 289–297.

Redekop, W.K., Koopmanschap, M.A., Stolk, R.P., Niessen, L.W., 2002.Health-related quality of life and treatment satisfaction in Dutchpatients with type 2 diabetes. Diabetes Care 25 (3), 458–463.

Sakane, N., Sato, J., Tsushita, K., Tsujii, S., Kotani, K., Tsuzaki, K., Tominaga,M., 2011. Prevention of type 2 diabetes in a primary healthcaresetting: three-year results of lifestyle intervention in Japanese sub-jects with impaired glucose tolerance. BMC Public Health 11, 40.

Shiu, A.T., Wong, R.Y., Thompson, D.R., 2003. Development of a reliableand valid Chinese version of the diabetes empowerment scale. Dia-betes Care 26 (10), 2817–2821.

Toobert, D.J., Hampson, S.E., Glasgow, R.E., 2000. The summary of diabetesself-care activities measure: results from 7 studies and a revised scale.Diabetes Care 23 (7), 943–950.

Wang, P.T., Hsieh, M.C., Hung, H.C., Ho, K.Y., Tsai, C.C., 2008. The associa-tion between periodontal condition and blood sugar control in type 2diabetic patient. Taiwan Journal of Oral Medical Sciences 24, 23–36.

Wong, A.K., Stewart, A.G., Furler, J.S., 2009. Development and validation ofthe diabetes management orientation scale (DMOS): assessing cul-turally related approaches to diabetes self-management. DiabetesResearch and Clinical Practice 86 (1), 24–30.

World Health Organization (WHO), 2012. Diabetes. In the World HealthStatistics Annual 2011. , http://www.who.int/features/factfiles/diabetes/en/index.html.