Nus Research Project

Master of Nursing MDN 5105: Research Project

HT043999E

Effectiveness of Brief Motivation Interviewing versus Health Education in Lowering Low-Density Lipoprotein levels in Newly Diagnosed Hyperlipidemia patients in the Community Setting 1. Introduction High Low Density Lipoprotein (LDL) levels are highly associated with acute coronary disease and ischaemic heart disease which are among the top 10 principle causes of death (Ministry of Health [MOH], 2005). According to MOH (2001) clinical practice guidelines, patients with LDL levels not within the normal range have to make lifestyle changes. The importance of changing health behavior thus reaching desired medical outcome cannot be undermine in this aspect. 1.1 Literature Reivew and Identification of Gaps Motiational Interviewing (MI) is a framework defined as “a directive, client-centered counseling style for eliciting behavior change by helping clients to resolve and explore ambivalence” (Rollnick and Miller, 1995 cited in Miller, 1996, p.839). The 5 basic principles include expressing empathy, developing discrepancy, avoiding argumentation, rolling with resistance and supporting self-efficacy. The human behavior theory that underlies MI is the Transtheorectical Change model (TTM). TTM believes that an individual undergo pre-contemplation and contemplation stages. These 2 stages are necessary before the individual reaches a stage to start preparing and taking action which resulted in a changed behavior. After a changed behavior, there are the maintenance phase and relapse phase. MI can be used in any of these stages (DiClemente and Velasquez, 2002). The main assumption of MI is that these stages of change are “fluid”, not as “fixed” in time as the TTM suggested. Thus it is possible to change a person’s stage of change by creating awarenesss of the ambivalence. Afterwhich, to explore this ambivalence and to resolve it. With a deeper understanding in human behavior, the usual health education, which consists didatic teaching, might not be effective in bringing about a change in human health behavior. Thus, employing brief MI counseling strategy can afterall change health

Effectiveness of Motivational Interviewing Study Page 1 of 19

Master of Nursing MDN 5105: Research Project

HT043999E

behaviors more effectively by exploring and resolving the ambivalence in one outpatient nurse consultation. Few studies had been done on Motivational Interviewing counseling technique on modifying dietary and exercise behaviors. Resnicow et al (2002) had identified only 6 studies done to measure the efficacy of MI in regards to modifying exercise and diet behaviors. 4 of these studies were secondary prevention trials and 2 of these studies were primary prevention trials. Out of these studies, only 1 study done by Mhurchu, Margetts and Speller (1998), studied on hyperlipidemia patients using MI based counseling in comparison to a standard dietary intervention. A meta-analysis on efficacy of MI (Burke, Arkowitz & Menchola 2003) revealed that the findings from 4 studies done using MI to modify diet and exercise behaviors remain preliminary. Furthermore, in most of these studies, the MI interventions were delivered by psychologists, dieticians and medical practitioners. Only the study by Woollard (1995) done on hypertensive patients had nurse counselors to deliver to the MI intervention. This paper proposes that MI can be modified into a brief sructured format which consists the 5 principles and with the same aim effectively change health behaviors in hyperlipidemia patients. The changing of health behaviors will result in better health medical outcomes. Also this brief structured format can used by registered staff nurses. This is further supported by the fact that MI was initially employed to change addictive behaviors. For nonaddictive behaviors, less time might be require to resolve ambivalence (Resnicow et al, 2002), thus the possibility of applying brief MI counseling strategy in outpatient setting to change health behaviors. Conclusions cannot be generalised to the general population as only 1 study had been done on hyperlipidemia patients. Futhermore, most of the studies were not done locally. MI a counseling strategy can be assumed as a method of communication. Different cultures have influences on the effectiveness or acceptability of a particular method of communication. Through doing this study in Singapore, acceptability and practicability

Effectiveness of Motivational Interviewing Study Page 2 of 19

Master of Nursing MDN 5105: Research Project

HT043999E

of this counseling strategy can be established by the subjects’ responses reported by the interventionist. Burke, Arkowitz and Dunn (2002) mentioned that the lack of integrity checks in the studies done resulted in low content validity of the independent variable of interest which is the MI intervention. Thus, weaknesses of present studies include the imbalances between internal validity and external validity of the studies. This study will be conducted in Singapore, at one of the branches of XXX Polyclinics, XXX polyclinic, as a pilot project. The full implementation of the study will be conducted across 9 polyclinics to cover the hyperlipidemia secondary prevention general population. This will help to establish external validity and conclusions of study can be generalised to the local population. The Brief MI intervention will have intergrity check to ascertain that the intervention given is MI and not other kinds of counseling strategy models like transtheoretic model (TTM) or cognitive behavior therapy (CBT), (Resnicow et al, 2002). 1.2 Significance of Study Changing health behaviors like changing eating habits and exercising regularly had been established as effective in lowering lipid levels. The lowering of lipid levels will in turn lead to a lower probability of having heart diseases and stroke. Although cost-anlaysis is not included in this study. It is logical to assume that costs of treating, rehabilitating and providing care to both heart diseases and stroke will be reduced with better medical outcomes. Lowering lipid levels not only have cost benefits but also other intangible benefits, e.g. better quality of life. Though in Singapore, nurses are already playing a crucial role in providing health education for hyperlipidemia patient in the current outpatient medical setting. It is important that health education or counseling continues to innovate and evolve. Using evidence based to establish the effectiveness of the current and “new” nursing interventions that can being given to change health behaviors.

Effectiveness of Motivational Interviewing Study Page 3 of 19

Master of Nursing MDN 5105: Research Project

HT043999E

1.3 Research Puporse/ Aim The aim of this study is to determine the effectiveness of Nurse-Led counseling strategy Brief Motivational Interviewing (MI) in lowering low-density lipoprotein (LDL) levels for newly diagnosed hyperlipidemia patients in outpatient medical setting. The conduction of the pilot study will at Yishun Polyclinic, one of the branches of National Healthcare Group Polyclinics (NHGP). The full conduction of this study will be done across the nine polyliclinics of NHGP. This research hypothesizes that Brief MI is more effective than the usual practice of health education in lowering LDL levels for newly diagnosed hyperlipidemia patients. 1.4 Ethical Considerations Ethical approval had been obtained from the DSRB committee on the 2nd September 2005 with reference no. DSRB-A/05/303 (Appendix A1). A sample of the consent form can be found on Appendix A2. 2. Methodology 2.1 Research Design A randomized prospective interventional study comparing effectiveness of lowering LDL levels using Brief MI counseling strategy versus Health Education in newly diagnosed hyperlipidemia patients in the primary healthcare setting. Objectives were to compare the effectiveness of the 2 methods using (1)primary outcome measures: Total cholestrol (TC), low-density lipoprotein (LDL), high-density lipoprotein (HDL), triglycerides (TG) and (2) secondary outcomes measures: dietary, exercise changes and body mass index (BMI). Differentiation of Interventions Given The usual health education involves didatic teaching and giving the patient information of good health behaviors. Brief MI counseling strategy is to used the 5 MI principles described above to move subject along the stage of change to want to change health behavior.

Effectiveness of Motivational Interviewing Study Page 4 of 19

Master of Nursing MDN 5105: Research Project

HT043999E

Integrity of the Brief MI counseling stratgy in this research will be addressed by structuring the counseling strategy in a “fixed” format as a conversational transcript (Appendix B1). Each question and statement structured had indication of the principle of MI that is being applied. Expert opinion will be sought to finalise the conversational transcript of the Brief MI intervention. The expert opinion committee consists of 3 senior nurse counselors with altogether over 80 years of experience in counseling patients and a National Healthcare Group health promotion doctor who specialise in teaching health education and counseling. A conversational transcript for health education is also formatted for the nurseinterventionist usage (Appendix B2). To reduce inter-rated reliability of the study, only 1 nurse will be the interventionist for this project. The nurse-interventionist is trained to follow the conversational transcript strictly in an afternoon session with a role-play session with the collaborator of this study. Questions were encouraged during the training session. This study had an initial plan to record the counseling session of each subject. After which, each member (blinded to the intervention carried out) of the expert opinion committee would assess individually the style of the counseling method used. The results will compare with the actual intervention given to further validate the difference and enhance the integrity of usual health education and Brief MI counseling strategy. However, this plan was unable to be carried out as most of the patients refused for their sessions to be recorded. 2.2 Sampling Plan Patients who fulfilled the following inclusion criteria will be recruited in this study. Inclusion Criteria include: •Newly diagnosed patients with hyperlipidemia not within normal range according to MOH clinical guidelines. •The initial diagnosis is made by polyclinic doctors •Subjects not prescribed with any anti-lipid lowering agents Effectiveness of Motivational Interviewing Study Page 5 of 19

Master of Nursing MDN 5105: Research Project

HT043999E

•Subjects with only primary hyperlipidemia will be included. •Subjects who are able to understand and converse in English or Mandarin will be recruited. This is because the MI counseling format will be written in English and Chinese format. •Subjects who are non-smokers. Smoking is an established risk factor for increased lipid levels and can decrease the potential effects of the intervention. Patients who are (a) newly diagnosed secondary hyperlipidemia cases (include patients with already a primary diagnosis of hypertension, diabetes mellitus etc.) (b) diagnosed by private clinics practitioners initially (c) unable to understand and converse in English and Mandarin (d) mentally confused and disorientated and (e) smokers will be excluded from this study. Potential subjects were referred by the doctor to the nurse interventionist. Sample of invitation letter can be found in Appendix C1. The nurse interventionist will recruit subjects according to the inclusion and exclusion criteria. The subjects will be randomized into 2 groups. Randomization is done using sealed envolopes method in batches of 10. After obtaining written consent, the nurse interventionist will draw a sealed envelop with instructions to employ the type of “intervention” to be given: usual health education or Brief MI counseling strategy. A flow chart on the recruitment process can be found in Appendix C2. As this project is a pilot study, limited manpower is allocated. Only 1 nurse interventionist is employed for this study. From Mhurchu, Margetts & Speller (1998) it is postulated that the MI group will have a further 10% reduction in all laboratory data (TC, LDL, HDL and TG levels) compared to the control group. Setting the standard deviation of the difference in all laboratory data between the two groups to be 10%, 20 subjects in each group will be required for a power of 80% and a 2-sided test of 5%.

Effectiveness of Motivational Interviewing Study Page 6 of 19

Master of Nursing MDN 5105: Research Project

HT043999E

2.3 Data Collection Methods Baseline data which include demographic data will be collected by the nurse prior to giving intervention to both groups. Primary outcome measures such as TC, LDL, HDL, and TG levels are analysed by National University Hospital laboratories from subjects’ blood taken by the NHG diagnostic laboratories. These readings are considered clinically reliable and valid for Singapore clinical context. Secondary data include dietary and exercise habits will be collected through questionnaires. Dietary habits consist of assessing (1) fat intake and (2) fruits and vegetables intake. After 3 months, subjects will return to the clinic for follow up, final post-intervention outcomes will be collected by the data collection nurse (another nurse different from the nurse interventionist). Dietary Fat Intake Questionnaire (Appendix D1) The Northwest Lipid Research Clinic a fat intake score is selected for this study due to its ease of self-administration, able to score under 3 minutes without a trained interviewer. The fat intake score is a brief dietary questionaire to screen and monitor diet intake related to plasma cholestrol (Retzlaff, Dowdy, Walden, Bovbjerg and Knopp, 1997). Reliability of this scale has a test-retest correlation coeeficients of 0.88 for men and 0.90 for women. Correlation with 4-day food records in total fat, saturated fat and cholesterol ranged from 0.42 to 0.60. These validity coefficients though lower than intensive instruments, however, they may be acceptable withing the outpatient clinical setting for general setting (Calfas, Zabinski and Rupp, 2000). As this questionaire is used in United Kingdom context, certain examples of food and units of measurement (e.g. ounces) are not acceptable in local context. NHGP dietician was consulted to replace the terms to local setting. Fruit and Vegetable Intake Questionnaire (Appendix D2) Intake of 2 servings of fruits and 2 servings of vegetables daily is recommended according to Ministry of Health (Health Promotion Board, 2005). The questionaire for fruit and vegetable intake is designed to ask subjects to specify the frequency and amount

Effectiveness of Motivational Interviewing Study Page 7 of 19

Master of Nursing MDN 5105: Research Project

HT043999E

of fruit and vegetable intake. NHGP dietician was consulted to verify the content validity of this questionnaire. Exercise Habits Questionnaire (Appendix D3) Internatonal Physical Activity Questionnaire (IPAQ), short version, was selected for this study to assess subjects’ exercise habits. The test re-test reliability coefficient for this instrument acorss European countries ranged from 0.67 to 0.95 (Oja, 2003). Total weekly activity in MET minutes for vigorous and moderate intensity activity and walking will be estimated using this instrument. Daily sitting time can be used as a measure of inactivity. 2.4 Data Analysis Plan All analyses will be performed using SPSS 13.0. 2 Sample T-test will be applied to compare the differences in the change between baseline and 3 month of LDL, HDL, TG and Total cholesterol when normality (checked using Komolgorov Smirnov 1 Sample test) and homogeneity assumptions are satisfied, otherwise Mann Whitney U test will be applied. A multiple regression analysis will be performed to adjust for relevant covariates*. McNemar test will be applied to assess the change in dietary and exercise habits with each group. Statistical significance will be set at p < 0.05. * As multiple regression analysis method had not been taught. This statistical analysis will not be performed for this study. This is approved by the Research Supervisor.

Effectiveness of Motivational Interviewing Study Page 8 of 19

Master of Nursing MDN 5105: Research Project

HT043999E

3. Results 3.1 Baseline analyses The usual care and intervention groups were compared on demographics, levels of lipid levels, body mass index (BMI) and fat score at baseline (Table 1 and 2). Within the Brief MI intervention group, 3 (75%) were male and 1 (25%) was female. The intervention group has a higher proportion of men compared to the usual care group which has 100% female subjects. All the subjects in this study are mainly Chinese; only 1 subject in the intervention group is an Indian. The usual care group appears to be less educated than the intervention group. All of them are primary level educated, whereas 50% of the Brief MI intervention group is educated at secondary level or higher. The mean age of subjects in the usual care group is 55, which is older than the Brief MI intervention group with a mean age of 49.5. Both groups have about the same mean BMI, 23.62 (control) and 23.94 (intervention), which falls above the health BMI range of 19.5 to 22.9. There are no statistically significant differences in the demographic characteristics, age and BMI between 2 groups. Baseline comparisons on the lipid levels and fat scores are done between 2 groups to establish similarities and potential confounding factors (Table 2). Generally the Brief MI intervention group has a poorer lipid profile compared to the usual care group. Higher mean TC levels (6.91mmol/L), LDL levels (4.30mmol/L), HDL levels (1.54mmol/L), and TG levels (1.44mmol/L) are reported in the intervention group. This is in comparison with the usual care group, with mean TC levels of 6.07mmol/L, LDL levels of 4.00mmol/L, HDL levels of 1.46mmol/L and TG levels of 1.34mmol/L. Again, statistically significant differences in these lipid levels cannot be established. Though the subjects in the intervention group have a poorer lipid profile, they reported better fat scores with a mean fat score of 26.75 compared to the usual care group with a mean fat score of 29.0.

Effectiveness of Motivational Interviewing Study Page 9 of 19

Master of Nursing MDN 5105: Research Project Table 1: Demographic Characteristics of Subjects Health Education Motivational Interviewing (n=2) (n= 4) Gender Male 3 (75%) Female 2 (100%) 1 (25%) Race Chinese 2 (100%) 3 (75%) Indian 1 (25%) Educational Level Primary 2 (100%) 2 (50%) Secondary 1 (25%) Tertiary 1 (25%) Mean Age ± SD 55 ± 4.24 49.5 ± 7.14 (52 to 58) (39 to 55) Mean BMI ± SD 23.62 ± 1.24 23.94 ± 2.21 (22.74 to 24.49) (20.88 to 25.65)

HT043999E

Sig. 0.40*

1.0*

0.47**

0.39** * 0.86** *

(a to b)= (min value to max value) * Using Chi-Square Fisher’s Exact Test of significance, p>0.05 ** Using Pearson Chi-Square Test of significance, p>0.0.5 *** Using Independent sample T-test of significance, p>0.05.Equal variances assumed, as Levene’s Test for equality of variances >0.05.

Table 2: Baseline Lipid Levels and Fat Scores between Motivational Health Education Interviewing (n=2) (n= 4) Lipid Levels Mean TC (mmol/L) ± SD 6.07 ± 0.35 6.91 ± 0.85 (5.82 to 6.32) (5.88 to 7.94) Mean LDL (mmol/L) ± SD 4.00 ± 0.32 4.30 ± 0.56 (3.78 to 4.23) (3.65 to 5.01) Mean HDL (mmol/L) ± SD 1.46 ± 0.07 1.54 ± 0.51 (1.41 to 1.51) (0.96 to 2.10) Mean TG (mmol/L) ± SD 1.34 ± 0.23 1.44 ± 0.42 (1.18 to 1.50) (0.88 to 1.84) Fat Score ± SD 29.0 ± 1.41 26.75 ± 1.71 (28 to 30) (25 to 29)

2 Groups Sig.

0.27+ 0.54+ 0.85+ 0.78+ 0.19+

(a to b)= (min value to max value) +:Using Independent sample T-test of significance, p>0.05. Equal variances assumed, Levene’s Test for equality of variances for all variables >0.05.

Effectiveness of Motivational Interviewing Study Page 10 of 19

Master of Nursing MDN 5105: Research Project

HT043999E

3.2 Follow-up data 3.2.1 Lipid Levels and Weight Table 3 contains the descriptive statistics as well as the difference scores from baseline to 3-month follow up for total cholesterol (TC), low-density lipoprotein (LDL), high-density lipoprotein (HDL), triglyceride (TG) and weight values in both groups. The subjects in the usual care group appear to perform better than the subjects in the intervention group. There was no statistically significant in the differences of improvements in the TC, HDL and TG levels between both groups. For LDL levels, the intervention group has increased LDL levels instead of a reduction. One of the subjects in the intervention group skewed the levels by having a post-LDL value of 5.13mmol/L with a pre-LDL value of 3.65mmol/L. When statistical tests were run without this subject, there was an improvement in LDL levels in the intervention group by 0.26mmol/L. However, this difference is not statistically significant with z score of -0.124, p-value 0.91. The intervention group has a reduction of weight in the subjects by 0.68kg (1.11% reduction) compared to the usual care group with an increased of 0.17kg. This difference is also statistically insignificant. When non-parametric is assumed for this analysis, Mann-Whitney U test of significance also showed no statically significant between the differences of mean rank score of TC, LDL, HDL, TG and weight between 2 groups (Table 4). The mean rank score for both TC and weight variables, 3.5, were similar for both groups. Although Table 4 showed that the usual care group did better than the intervention group in LDL, HDL and TG values, the mean rank score differences range from only 0.75 to 1.5.

Effectiveness of Motivational Interviewing Study Page 11 of 19

Master of Nursing MDN 5105: Research Project

HT043999E

Table 3: Mean values of Lipid Levels and Weight changes between Baseline and 3-month follow-up Health Motivational Test statistic P-value+ Education Interviewing TC Levels Baseline M ± SD 6.07 ± 0.35 6.91 ± 0.85 t = -1.28 0.27 Post M ± SD 5.84 ± 0.34 6.74 ± 0.78 t = -1.48 0.21 Δ score ± SD -0.23 ± 0.69 -0.17 ± 0.46 z = -0.13 0.91 % Reduction 3.79% 2.46% LDL Levels Baseline M ± SD 4.00 ± 0.32 4.30 ± 0.56 t = -0.67 0.54 Post M ± SD 3.68 ±0.36 4.47 ± 0.52 t = -1.90 0.13 Δ score ± SD -0.33 ± 0.68 0.17 ± 0.98 z = -0.64 0.56 % Reduction 8.25% -3.95% HDL Levels Baseline M ± SD 1.46 ± 0.07 1.54 ± 0.51 t = -0.20 0.85 Post M ± SD 1.60 ± 0.06 1. 61 ± 0.53 t = -0.04 0.97 Δ score ± SD 0.14 ± 0.18 0.07 ± 0.14 z = 0.52 0.63 % Reduction 9.59% 4.54% TG Levels Baseline M ± SD 1.34 ± 0.23 1.44 ± 0.42 t = -0.30 0.78 Post M ± SD 1.23 ± 0.08 1.42 ± 0.14 t = -1.78 0.15 Δ score ± SD -0.12 ± 0.30 -0.02 ± 0.29 z = -0.38 0.72 % Reduction 8.96% <0.01% Weight Baseline M ± SD 54.50 ± 4.24 61.15 ± 11.75 t = -0.74 0.50 Post M ± SD 55.4 ± 2.69 60.47 ± 11.22 t = -0.60 0.58 Δ score ± SD 0.9 ± 1.56 -0.68 ± 0.64 z = 1.91 0.13 % Reduction -0.17% 1.11% + Equal variances assumed, Levene’s Test for equality of variances for all variables >0.05.

Table 4: Differences in Rank Scores1 of Changes between Baseline and 3-month follow-up HE+ MI+ P-value.+ Mean Sum of Mean Sum of Rank Score Ranks Rank Score Ranks TC 3.5 7 3.5 14 1.0 LDL 4.5 9 3.0 12 0.53 HDL 4.25 8.5 3.13 12.5 0.53 TG 4.00 8 3.25 13 0.8 Weight 3.5 7 3.5 14 1.0 1. Scores are ranked in order of 1 – lowest degree of change to 6 – highest degree of change + Using Mann-Whitney U Test of Significance.

Effectiveness of Motivational Interviewing Study Page 12 of 19

Master of Nursing MDN 5105: Research Project

HT043999E

3.2.2 Fat Dietary Intake A score of 24 or less in the Northwest Lipid Research Clinic fat intake score indicates a diet intake of moderate to low fat and cholestrol. Table 5 shows the fat intake scores for both at baseline and follow up were above 24 for the usual care group (29, 26.5), although there was a 2.5 reduction in the score. The intervention group has an improvement of 3.5 score rating from 26.75 to 23.25. Statistically significane in the changes of fat score between these 2 groups cannot be established. Table 5: Mean values of Fat Score changes between Baseline and 3month follow-up Health Motivational Test statistic P-value+ Education Interviewing Baseline M ± SD 29.00 ± 1.41 26.75 ± 1.71 t = 1.59 0.19 Post M ± SD 26.50 ± 3.54 23.25 ± 2.22 t = 1.44 0.22 Δ score ± SD -2.5 ± 4.95 -3.5 ± 2.38 z = 0.36 0.74 Fat score less than 26 considered adherence to low fat diet. + Equal variances assumed, Levene’s Test for equality of variances for all variables >0.05.

3.2.3 Fruit and Vegetable Intake Table 6 and 7 show the changes in fruit intake frequencies and amount in the 2 groups at baseline and 3-month follow up. Subjects in both groups showed improvements at 3month follow up, reaching the ideal fruit intake frequency. For fruit amount intake, 1 subject (50%) in the usual care group did not reach the adequate fruit amount intake of 2 servings after usual care intervention. Whereas the subjects in the brief MI counseling group all the subjects maintain/ reach adequate fruit amount intake. The differences in this change of fruit amount intake are insignificant using McNemar test of significance. Table 6: Changes in Fruit Intake Frequencies in 2 Groups between baseline and 3-month follow-up Health Education Post-Intervention Fruit Intake Frequency Pre-Intervention Fruit Intake Frequency Ideal Non-ideal Ideal1 1a 0 Non-ideal2 1a 0 Motivational Interviewing Post-Intervention Fruit Intake Frequency Pre-Intervention Fruit Intake Frequency Ideal Non-ideal a Ideal 2 0 Non-ideal 2a 0

Effectiveness of Motivational Interviewing Study Page 13 of 19

Master of Nursing MDN 5105: Research Project

HT043999E

1. Ideal: Consume fruits daily. 2. Non-ideal: Consume fruits less than daily. a: p>0.05 using McNemar Test of Significance for both groups

Table 7: Changes in Fruit Amount Intake in 2 Groups between baseline and 3-month follow-up Health Education Post-Intervention Fruit Amount Pre-Intervention Fruit Amount Adequate Not Adequate Adequate 1 1a 0 2 Not Adequate 0 1a Motivational Interviewing Post-Intervention Fruit Amount Pre-Intervention Fruit Amount Adequate Not Adequate Adequate 2a 0 Not Adequate 2a 0 1. Adequate: Consume 2 or more servings of fruits 2. Not Adequate: Consume less than 2 servings of fruits 1 serving of fruit = a: p>0.05 using McNemar Test of Significance for both groups

All the subjects reported daily intake of vegetables at baseline for both groups. This fulfills the recommendations by the Ministry of Health. However, 1 subject (50%) in the usual care group and 2 subjects (50%) from the intervention group do not fulfill the adequate vegetable amount intake. There was no change of vegetable amount intake after usual care intervention in this subject, whereas the 2 subjects from the brief MI counseling group reach the adequate vegetable amount intake at 3-month follow up. These differences were also not statistically significant. Table 8: Changes Vegetable Amount Intake in 2 Groups between baseline and 3-month follow-up Health Education Post-Intervention Veg Amount Pre-Intervention Veg Amount Adequate Not Adequate 1 a Adequate 1 0 Not Adequate 2 0 1a Motivational Interviewing Post-Intervention Veg Amount Pre-Intervention Veg Amount Adequate Not Adequate Adequate 2a 0 a Not Adequate 2 0 1. Adequate: Consume 2 or more servings of vegetables. 2. Not Adequate: Consume less than 2 servings of vegetables. 1 serving of vegetables = a: p>0.05 using McNemar Test of Significance for both groups

Effectiveness of Motivational Interviewing Study Page 14 of 19

Master of Nursing MDN 5105: Research Project

HT043999E

3.2.4 Exercise Habits Estimation for total weekly activity in MET minutes for vigorous and moderate intensity activity and walking was supposed to be done. However, the subjects for this study had difficulty recalling the amount of time they spent on physical activity. Thus, only frequencies for intensive and moderate activities were captured. All the subjects in the 2 groups did not engage in some form of intensive activities at baseline or 3-month follow up. Changes in engagement of moderate activities were present and shown in Table 9. 1 subject in the usual care group did not engage in moderate activities at baseline. This health behavior did not change at 3-month follow up. For the intervention group, 3 subjects whom did not engage in moderate activities, engage in some form of activities at 3-month follow up. The remaining only subject in the intervention group whom did engage in moderate activities at baseline, reported not engaging in moderate activities at 3-month follow up. Tables 9: Changes in engagement of Moderate Activities in 2 Groups between baseline and 3-month follow-up Health Education Post-Intervention Moderate Activities engagement Pre-Intervention Moderate Activities engagement1 Yes No Yes 1a 0 No 0 1a Motivational Interviewing Post-Intervention Moderate Activities engagement Pre-Intervention Moderate Activities engagement Yes No Yes 0 1a No 3a 0 1. Moderate exercise - activities that take moderate physical effort and make you breathe somewhat harder than normal. E.g. cycling at regular pace, swimming, etc. Do not include walking a: p>0.05 using McNemar Test of Significance for both groups.

Effectiveness of Motivational Interviewing Study Page 15 of 19

Master of Nursing MDN 5105: Research Project

HT043999E

4. Discussion Lifestyle changes counseling or education for people with hyperlipidemia is based on 3 assumptions: education increases knowledge, increased knowledge leads to an improvement in dietary and exercise habits and finally the changed health habits lead to a reduction in the lipid levels. Even though the results suggested that the brief MI counseling strategy group did better in the fat score, fruit and vegetable amount intake and engaging in moderate exercise compared to the usual care group. These also suggest that brief MI counseling intervention can change health behaviors. The fact that statically significance cannot be established, brief MI counseling strategy cannot be viewed as more effective or superior to the usual health education. Although it can be ascertain that brief MI counseling strategy is not less effective or inferior to the usual health education. Furthermore, evidence for MI counseling strategy in the area of diet, exercise and other lifestyle behaviors is mixed (Burke, Arkowitz and Dunn, 2002). The findings in this study are similar to the Mhurchu, Margetts and Speller’s (1998) study on hyperlipidemia patients with MI as the intervention studied. There was also no significant difference in the reductions of lipid levels between the 2 groups. The timing between the effects of changed health behaviors on the lipid levels is an important consideration for this study. This study assumed that the effects of changed health behaviors will happen at 3 month. However there is this possibility that the intervention group after changes in health behaviors might lead to decrease in the lipid levels at a later time or more sustained reduction of lipid levels which cannot be established in this study than the usual care group.

Effectiveness of Motivational Interviewing Study Page 16 of 19

Master of Nursing MDN 5105: Research Project

HT043999E

5. Limitations The original calculated sample size of 40 (to obtain 80% power) was not achieved and this reduced the power of the study. The reduction in study power suggests that results should be interpreted with some caution since it is possible that the smaller sample size may have reduced the power to detect as statistically significant difference. Few factors contribute to the small sample size: (1) too stringent inclusion and exclusion criteria (2) small incidence of pure hyperlipidemia patients in Yishun polyclinic. Health behaviors can be influenced by a lot of factors that are uncontrolled in the clinical setting. One subject in the intervention group which had extremely post high LDL value reported taking daily evening primrose pills to bring down her cholesterol. The subject in the intervention group which pre-intervention engage in moderate activities stopped engaging in moderate activities due to a change of his work nature. Lastly, one of the subjects in the usual care group is the wife of another subject in the intervention group. Although both of them have reductions in the LDL values and improvement in health habits, the changes due to the interventions cannot be ascertain as family support might be factor in resulting in these changes. 6. Conclusion In conclusion, there are evidences to suggest brief MI counseling strategy can change health behaviors. The small sample size is the main concern for this study results. The study should be done in a larger scale. Time period for assessment in a change of health behaviors for the next project can be done at a more regular period of 1 week post intervention, 3 months and 6 months. The lipid levels can also be considered to measure also at a 6-month period to determine the effects of the change health behaviors. Lastly, subjects who lived in the same household should be captured and analysis can be adjusted accordingly.

Effectiveness of Motivational Interviewing Study Page 17 of 19

Master of Nursing MDN 5105: Research Project

HT043999E

Brief MI counseling strategy, at preliminary stage, showed that there might be used in changing health behaviors. However, the effectiveness of this strategy cannot be established as more superior than the usual health education. But it can be ascertain that it is not less inferior to usual health education.

Effectiveness of Motivational Interviewing Study Page 18 of 19

Master of Nursing MDN 5105: Research Project

HT043999E

References Burke, B.L., Arkowitz, H. and Dunn, C. (2003). The efficacy of motivational interviewing and its adaptations: what we know so far. In W.R. Miller and S. Rollnick, Motivational interviewing – preparing people for change, pp. 217 - 250. New York: The Guilford Press. Burke, B.L., Arkowitz, H. and Menchola, M. (2003). The efficacy of motivational interviewing: a meta-analysis of controlled clinical trials. Journal of Consulting and Clinical Psychology, 71(5), 843 – 861. Calfas, K.J., Marion, F.Z. and Rupp, J. (2000). Practical nutrition assessment in primary care settings – a review. American Journal of Preventive Medicine, 18(4), 289 – 299. DiClemente, C.C. and Velasquez, M.M. (2002). Motivational interviewing and the stages of change. In W.R. Miller and S. Rollnick, Motivational interviewing – preparing people for change, pp. 201 – 216. New York: The Guilford Press. Health Promotion Board, (2005). Education materials – health tips. Retrieved on 27 July 2005 from http://www.hpb.gov.sg/hpb/default.asp?pg_id=991 Mhurchu, C.N., Margetts, B.M. & Speller, V. (1998). Randomized controlled trial comparing the effectiveness of 2 dietary interventions for patients with hyperlipidemia. Clinical Science, 95, 479-487. Miller, W.R. (1996). Motivation interviewing: research, practice and puzzles. Addictive Behaviors, 21(6), 835 – 842. Oja, P. (2003). Development of a common insturment for physical activity. In A. Nosikov and C.Gudex, EUROHIS: developing common instruments for health surveys. IOS Press. Ministry of Health, Singapore (2001). Clinical practice guidelines – lipids. Singapore Ministry of Health. Ministry of Health, Singapore (2005). Principal causes of death. Retrieved on 28 March 2006 from http://www.moh.gov.sg/corp/publications/statistics/principal.do Reniscow, K., Dilorio, C., Soet, J.E., Borrelli, B. and Hecht, J. (2002). Motivational interviewing in health promotion: it sounds like something is changing. Health Psychology, 21(5), 444 – 451. Retzlaff, B.M., Dowdy, A.A., Walden, C.E., Bovbjerg, V.E. and Knopp, R.H. (1997). The northwest lipid research clinic fat intake scale: validation and utility. American Journal of Public Health, 87(2), 181 – 185.

Effectiveness of Motivational Interviewing Study Page 19 of 19