Ronnachai Kongsakon, M.D.,LL.B.,M.Sc.*
Teerakiat Jareonsettasin, M.D. MRCPsych **

Objective The objective of this study was to assess the reliability and validity of the Client Satisfaction Questionnaire (CSQ-8) Thai Version in 100 patients with known mental disorders. Study at Outpatient and Inpatient Departments of Psychiatry, Ramatibodhi Hospital, Bangkok, Thailand.

Method Cross–sectional study. A Thai version of the CSQ-8 was developed through forward-backward translation techniques and reviewed by five content-experts using psychometric methods to test the reliability and validity of the version. Main outcome measuresare patient-satisfaction and scores for specific aspects of satisfaction. Relation between satisfaction and patient's sex and type of mental disorder.

Results An eight-item questionnaire was developed. The validity with the item coefficient >0.6 for all scales.The reliability of the CSQ-8 Thai Version on a test and retest had a Pearson’s correlation of 0.697 and a Cronbach's Alpha coefficient for all scales of >0.70. The scale’s coefficient for the total number of items was 0.7897. Conclusion Our study supports the use of the CSQ-8 Thai Version as a reliable guide in the Thai setting to most psychiatric patient’s satisfaction.

J Psychiatr Assoc Thailand 2000; 45(2):155-163.

Key words: outcome assessment, reliability, validity, Client Satisfaction Questionnaire (CSQ-8) Thai Version, satisfaction, mental

** Department of Psychiatry, Faculty of Medicine, Khon Kaen University, Khon Kaen, 40002.

วัตถุประสงค์ การศึกษานี้ต้องการตรวจสอบความเที่ยงตรงและความแม่นยำของแบบวัดความพึงพอใจของผู้รับบริการ(CSQ8)ฉบับภาษาไทยในผู้ป่วยจิตเวชจำนวน100คน

วิธีการศึกษา ศึกษาเชิงพรรณาแบบตัดขวาง โดยแบบวัดความพึงพอใจฉบับภาษาไทยได้ผ่านขบวนการแปลจากฉบับภาษาอังกฤษ โดยนักภาษาศาสตร์ และแปลกลับเป็นภาษาอังกฤษ ประเมินโดยผู้เชี่ยวชาญด้านเนื้อหา 5 คน มาทดสอบโดยวิธีการตรวจสอบคุณสมบัติแบบวัด ในเรื่องความเที่ยงตรงและความแม่นยำ โดยมีตัววัดที่สำคัญ คือ คะแนนความพึงพอใจของผู้ป่วย ความสัมพันธ์ระหว่างความพึงพอใจกับเพศ และความผิดปกติของผู้ป่วย

ผลการศึกษา แบบสอบถามความพึงพอใจจำนวน 8 ข้อ ผ่านการตรวจสอบ โดยมีค่าความเชื่อถือได้ ของค่า test-retest Pearson’s correlation = 0.697, Cronbach’s alpha coefficient = 0.789 ความแม่นตรงจากการแปลและตรวจสอบโดยผู้เชี่ยวชาญ มีค่า item coefficient > 0.6

สรุป การศึกษาในครั้งนี้สนับสนุนการใช้ แบบวัดความพึงพอใจของผู้รับบริการ( CSQ-8) ฉบับภาษาไทยในการประเมินความพึงพอใจของผู้ป่วยต่อการรักษาในประเทศไทยด้วยตัวผู้ป่วยเอง

วารสารสมาคมจิตแพทย์แห่งประเทศไทย 2543; 45(2): 155-163.

คำสำคัญ การประเมินผล ความเชื่อถือได้ ความแม่นตรง แบบวัดความพึงพอใจผู้รับบริการ CSQ-8 ฉบับภาษาไทย ความพึงพอใจ, ความผิดปกติทางจิต

** ภาควิชาจิตเวชศาสตร์ คณะแพทยศาสตร์ มหาวิทยาลัยขอนแก่น อำเภอเมือง จังหวัดขอนแก่น 40002

Reforms to health care system have increased pressure on healthcare providers and users to monitor patient satisfaction. Perceived quality of service is multifaceted and subjective so multiple probes are required to adequately characterize it. Response times, telephone advice rates, prescriptions, admission rates, and patient’s health status and perceived satisfaction all must be taken into consideration. Low patient satisfaction may result in poor compliance (a waste of resources) and a suboptimal clinical outcome¹. Satisfaction of the legitimate demands of patients is therefore an objective of all medical care and should be included as an outcome measure.

Though measuring satisfaction of medical care is not straightforward, patients' views can be explored and brought to bear for the improvement of healthcare². One approach is to use qualitative methods³, but these are difficult to use for routine large scale service evaluations. An alternative is to use a quantitative questionnaire, where the random error of responses is minimised so that consistency of measurement is achieved. The questionnaire must be validated so that there is confidence it truly reflects what it seeks to measure⁴ .Validity of questionnaires can be assessed superficially, by content, criterion, or construct validity^5,6.

There is disenchantment with structured questionnaire surveys as appropriate instruments because of the difficulty of ensuring adequate coverage, a high response, and reliability of the questions⁷.Correction by surveying overly heterogeneous patient populations may generate irrelevant information to the needs of specific client groups. Much of the early work measuring patient satisfaction took place in the United States but similar such work has not been published in Thailand. There is therefore a need to clarify and validate a questionnaire pertinent for the Thai situation.

The client satisfaction questionnaire CSQ^8,9 is a self-report questionnaire constructed to measure satisfaction with services received by individuals and families. It has been adopted internationally by investigators because it has good psychometric properties. The alpha coefficient for the CSQ-8 is 0.93, indicating that it possesses a high degree of internal consistency and its eight items provide a homogeneous estimate of general satisfaction with services received.

We report on the development of a Thai version of the CSQ-8 for measuring patient satisfaction with psychiatric care in Thailand, which can be administered either by interview or completed by patient s or their caregiver.

Steps in the development of a the CSQ-8 Thai Version:

1. translated into Thai,

2. back-translated by a bilingual (Thai-English) speaker to judge whether the meaning matched the

original English version, and

3. tested for validity and reliability.

The study was undertaken with the assistance of four psychiatric nurses at the out- and inpatient, Departments of Psychiatry of Ramatibodhi Hospital between April 1 and June, 30, 1999. with the sample size containing at least 100 patients was large enough to detect a difference in the satisfaction score of a quarter of a standard deviation at a significance level of 0.05 and 80% power.

The sample consisted of patients with mental disorders. The nurses distributed information detailing the objectives of the study to all incoming patients. Entry criteria were that the patient gave consent and had presented symptoms of mental illness for more than seven days. If the patient was eligible, baseline data were requested and the patient was given the CSQ-8 Thai Version for scoring (demographic data are presented in Table 1.). A follow-up was posted two weeks later.

Data were analyzed using SPSS for Windows¹⁰. Questions were scored from ‘1’ to ‘4’, where ‘4’ represented strong agreement with a statement of satisfaction and ‘1’ represented strong disagreement. For negatively worded items, the scores were reversed so that a high score always indicated satisfaction. The score for each patient on each scale of satisfaction represented their average response to the questions that made up that scale.

Back translation of the Thai version of the questionnaire gave nearly the same meaning as the original English text, so we have some assurance that the psychometric properties of the scale have been retained through translation.

To verify the adequacy of the questionnaire’s probes in the Thai cultural context, the Thai version CSQ-8 was sent to content-experts (three psychiatrists and two social scientists). They determined the coefficient for all items were >0.6.

There was no statistical difference in gender with the mean score by student’s t test, the satisfaction of male and female to the service was not different as shown in Table3.

Comparision of the mean score of different type of disorder by ANOVA, there was no statistical difference as shown in Table 4.

The reliability coefficient tests were performed using SPSS Version 7.5 for Windows¹⁰The alpha coefficients are shown in Table 5.

Using a split-half analysis, the alpha for part 1 was .6277 and .7308 for part 2, which are well correlated indicating a satisfactory degree of reliability.

The stability is the reproducibility of a measure administered on different occasions. As this questionnaires are self-administered , the test retest reliability is the only method to test for stability.

The mean scores and Pearson’s correlation of the test and retest score after two weeks apart comes out; as shown in Table 6. The Pearson’s correlation of the test and retest (two weeks later) was 0.697( p<0.001)

These findings indicate that the questionnaire was reliable and valid and therefore suitable for evaluating the psychiatric care received by a broad range of mental patients.

According to Cronbach's coefficients which were greater than 0.70 for all items, the questionnaire had good internal reliability. The test and retest scores were highly correlated, though the regressions showed that the retest scores were lower. Since the variable and measurement technique should be the same on both occasions, the lower retest scores may reflect a difference in the method of application, with greater satisfaction being expressed when the research assistants were present. Alternatively, there may have been a real fall in satisfaction with time. Nevertheless, these data indicate that the retest reliability of the questionnaire was satisfactory.

Content validity was ensured by the process of questionnaire development. Issues important to patients were identified during the qualitative phase and included with elements identified from other studies ⁸ then given to five experts who assessed the acceptable items and found high coefficients of acceptability. This Thai Version of the CSQ had no problems with content validity and no culturally induced obfuscations. The inter-item correlations show that though each item was correlated with and hence related to overall satisfaction, the scales assessed different aspects of satisfaction and contributed to a global measure of satisfaction.

The acceptability of the questionnaire to patients was revealed by the high response rates (>50%, Table 6) to the posted retest and the high proportion of responses for which we could calculate scale scores. This suggests that the instrument may successfully be administered by post to a broad range of urban psychiatric patients.

The range of scores obtained (19 to 32, Table 2) confutes the possibility that the questionnaire forced or skewed responses and confirms that the questionnaire can detect differences in satisfaction.

Comparisons of the means of sex and types of mental disorders (Tables 3 and 4) yielded no statistically significant differences perhaps because the group was from one setting and there really was no measurable difference in the degree of service.

New models of healthcare are being developed and both new and existing models should be evaluated or audited to ensure that the needs of both patients and practitioners are being met. This will require measurement of patient satisfaction. Overall evaluations will require a judgment of the relative importance of each need, but no need of any, who use, provide, or pay for the service being evaluated, should be ignored. We therefore encourage all who wish to evaluate service to include an assessment of patient satisfaction.

We have developed a questionnaire (the CSQ-8 Thai Version) which was acceptable to the mental patients tested, and proved its reliability and validity. This instrument can be used for outcome assessment with most psychiatric patients in the Thai setting. Further development of the questionnaire for other settings and other types of patient populations would be desirable and feasible.

The authors thank for the Grant support from Les Laboratoires Servier and Mr. Bryan Roderick Hamman for his assistance in developing the English-language presentation of the paper.

Reference

Table 2. The mean score of the samples with the CSQ-8 Thai Version

Table 3. Comparison of the means by sex

๑. ท่านจะประเมิน คุณภาพของการบริการที่ท่านได้รับอย่างไร

๔ ๓ ๒ ๑

ดีมาก ดี ปานกลาง ไม่ดี

๒. ท่านได้รับสิ่งที่ท่านต้องการมากน้อยเพียงใด

๔ ๓ ๒ ๑

ไม่ได้รับเลย ไม่ได้รับมาก ได้รับทั่วๆไป ได้รับตามความต้องการทั้งหมด

๓. การรักษาที่ผ่านมา ตรงกับความต้องการของท่านอย่างไร

๔ ๓ ๒ ๑

ตรงกับความต้องการทั้งหมด ตรงเกือบทั้งหมด ตรงเพียงบางส่วน ไม่ตรงกับความต้องการเลย

๔. ถ้ามีเพื่อนที่มีปัญหาเหมือนท่าน ท่านจะแนะนำให้มารักษาที่นี้หรือไม่

๔ ๓ ๒ ๑

ไม่แนะนำแน่นอน ไม่คิดจะแนะนำ คิดว่าคงจะแนะนำ แนะนำอย่างแน่นอน

๕. ท่านรู้สึกพึงพอใจในปริมาณความช่วยเหลือที่ท่านได้รับความช่วยเหลืออย่างไร

๔ ๓ ๒ ๑

ไม่พอใจ เฉยๆหรือไม่พอใจบ้าง ส่วนใหญ่พอใจ พอใจมาก

๖. สิ่งที่ท่านได้รับความช่วยเหลือ สามารถช่วยท่าน แก้ไขปัญหาได้อย่างมีประสิทธิภาพ

มากขึ้น หรือไม่

๔ ๓ ๒ ๑

ช่วยได้อย่างมาก ช่วยได้บ้าง ไม่เลย ไม่ช่วยและรู้สึกทำให้ปัญหาแย่ลง

๗. ในภาพรวม ทั่วๆไป ท่านรู้สึกพอใจกับการบริการรักษาที่ผ่านมาอย่างไร

๔ ๓ ๒ ๑

พอใจมาก ส่วนใหญ่พอใจ เฉยๆ หรือ ไม่พอใจเล็กน้อย ไม่พอใจ

๘. ถ้าท่านต้องการความช่วยเหลืออีก ท่านจะกลับมารักษาที่นี่หรือไม่

๔ ๓ ๒ ๑

ไม่มาแน่นอน คิดว่าไม่มา คงจะมา มาแน่นอน