Development of Audit Tool: A methodological study for Auditing the Family Health Records

Transcription

1 Development of Audit Tool: A methodological study for Auditing the Family Health Records Bandana, Indarjit Walia, Sushma Kumari Saini Abstract :The Audit is the major Tool used to evaluate quality of care. The Audit Tool is developed to audit the family health records. The Audit Tool is developed in six phases. In first phase, the preliminary draft of Audit Tool is prepared.in second phase, content validity was checked with the help of Delphi technique. In third phase, necessary modifications were done in the Tool. In fourth phase, feasibility of the Tool is assessed by doing a pilot study. In fifth phase, final draft of the Tool was prepared. In sixth phase, auditing of 500 family folders was done to check the content validity and construct validity of the Tool by applying Cronbach's alpha and principal component factor analysis method. Cronbach's alpha was 0.73, which shows that Tool was reliable. The Kmo value was and p-value was 0.000, which makes the data eligible for factor analysis. There were 22 items in the original Tool, but after factor analysis 21 items are retained. This study has made contribution to evaluate the family health record maintained by the nursing students of NINE, PGIMER, Chandigarh with minimum time. Besides this, it will also help in improving the quality of documentation. This Tool enables the supervisors /faculty not to cross check the record by visiting the families personally as this Tool is self explanatory. Key words : Audit tool, Family health records, Nursing students Correspondence at : Mrs. Bandana Lecturer Shri Guru Harkrishan College of Nursing, Sohana Introduction Quality is the process of meeting the needs and expectations of customers, both internal and external. 1 Quality can also be referred to as a continuous process of incremental improvement. Quality is never an accident; 2 it is always the result of high intention, sincere effor t, intelligent direction and skillful execution. 3 It represents the wise choice of many alternatives. 166

2 Audit is a kind of review & is a major tool used to evaluate quality of care 4. An audit is the systematic review of records for the purpose of evaluating the documentation. 5 traditionally, audits were mainly associated with gaining information about financial systems and the financial records of a company or a business. An audit involves examination and comparison of results against set criteria; assessment of quality of care with a view to improve; and monitoring the effect of audit-induced changes. 6 Nursing audit is the assessment of the quality of nursing care &uses a record as an aid in evaluating the quality of patient care. Nursing audit refers to assessment of the quality of clinical nursing. The audit is a means by which nurses can define standards from their point of view and describe the actual practice of nursing 7. The first report of a nursing audit of a hospital was published in The purposes of a nursing audit are to evaluate the given nursing care, to check the quality of nursing care; to stimulate the better recording, to focus on care provided and not on care provider; and to contribute to research 8. A nursing audit is of two types. First, a retrospective audit which refers to an in-depth assessment of the quality after the patient has been discharged patient s chart is the source of data. A retrospective audit is a method for evaluating the quality of nursing care as it is reflected in the patient care records for a discharged patient. In this type of audit, specific behaviors are described and then converted into questions, and the examiner looks for answers in the record 9. The second is a concurrent audit which refers to evaluation when a patient is still undergoing care. It includes assessing the patient at the bedside in relation to predetermined criteria; interviewing the staff responsible for this care; and reviewing the patient s record and care plan. 10 Today, one of the most challenging issues in nursing practice is how to document quality client care within the constraints imposed by regulation, resources and finances. Till now, no single documentation system has been pointed out as effective; each one has its own limitations. Family is the basic unit of health services. Family folders are the documents in which the health records are maintained by BSc Nursing (4years), BSc Nursing (Post- Basic) and MSc Nursing students of National Institute of Nursing Education, Post Graduate Institute of Medical Education and Research Chandigarh, have been visiting rural areas and semi-urban areas of the Union Territory of Chandigarh for many years. They have been providing preventive, promotive and curative services to the above mentioned communities. They have been visiting families regularly and these visits are recorded in a family folder. There is a single folder for one family, which can also be called family health record. There is no Tool to measure the completeness of the components of a family health record without going in the family itself. So, there is need for an instrument to audit such records. Hence a study was conducted to develop a tool for auditing these records. 167

3 This will contribute to evaluate the family health record maintained by the nursing students with minimum time. Besides this, it will also help in improving the quality of documentation. Objective Main objective of the study was to develop a Tool for auditing the family health records maintained by nursing students of NINE, PGIMER, Chandigarh. Materials, Methods and Results The study was conducted on the health records of people of Dadu Majra Colony. This colony is situated on the north-west corner of Chandigarh and is at a distance of 5 km from NINE, PGIMER Chandigarh and 10 km from Inter-State Bus Terminus sector-43 Chandigarh. Dadu Majra Colony a resettled colony of UT Chandigarh. It has 2,670 houses with a population 18,000 approximately. The residents of colony are migrants from different states of India, including Punjab, Haryana, Himachal Pradesh, Rajasthan, Utter Pradesh, Bihar and Uttrakhand. The socio-economic status of residents ranges from lower to upper middle class with different cultural and social practices. Dadu Majra Colony is chosen purposely for the study because students of the NINE, PGIMER, Chandigarh are providing their preventive, promotive and curative services to the above mentioned communities. They visit the families regularly and these visits are recorded in a family folder. There is a single family folder for an individual family, which is also called the family health record. On the front page of family folder name of the head of the family, Address, Caste, Occupation & Sources of family income are entered. On the back page of family folder date, family characteristics and environment are entered. Inside the family folder the postnatal observation, neonatal card, child under-six, women s record including obstetric history & antenatal record are attached Audit Tool was developed in six phases. The first phase was preliminary phase. This phase was completed in four steps. In the preliminary phase, components of health record tool were determined and item pool was generated and organized. The preliminary draft of the audit Tool was prepared. In the second phase, content validity was checked with the help of Delphi technique. Nine Delphi rounds were done. Eleven experts from the field of nursing education & eight experts from the department of school of public health have been consulted and requested to check the face validity, item validity, item wording, and item order. Nine drafts were prepared in stages. In every stage, experts opinion has been consulted and after incorporating their valuable suggestions modifications were done in the tool. The first draft was the preliminary draft. In the second draft, there were five columns (i.e. serial number (SN), comparison of health record with family, complete, incomplete and wrong.) There was scoring for every item and grading as poor, average, good, very good and excellent. The third draft was prepared in two parts. In the first part, items whose number varies from family to family (women s record, 168

4 antenatal record, obstetrics record and child under six records) were included. In the second part of this draft, a total 19 items were included. In the fourth draft, par t 1 was named as part A and part 2 was named as par t-b. In the fif th draf t, the sociodemographic profile term was replaced by the terms identification and social status. In this draft, the items (women s record, antenatal record, obstetrics record and child under six records) were arranged as T/O (total score/ obtained score). The sixth draft was prepared in two parts. In part A, main items related to maintenance of quality were included, and in part-b numbers 1, 2, 3, were replaced by A, B, C. In the seventh draft, in part A, there were three columns and for part B the items related to quality of maintenance of family health record were included. In the eighth draft, all the items were categorized under three headings: writing style (total items 12), visit details (total items 5) and record keeping (total items 2). In the ninth draft, a total 10 components were constr ucted. These components were family background, maintenance of quality, use of nursing process, family characteristics, women record, antenatal record, obstetrics record, postnatal record, neonatal record and child under six records. In the third phase, the necessary modifications were done in the tool under the guidance of guide and co- guide. In the fourth phase, the feasibility of the tool is assessed by doing a pilot study on family health records of village Dhanas, Chandigarh. Village Dhanas was chosen for the pilot study because the sociodemographic profile of the people of Dhanas is similar to the people of Dadu Majra Colony and same students are maintaining the records in both areas. In the pilot study the Tool was found feasible. In the fifth phase, the final draft of Audit Tool was prepared by incorporating the minor changes after the pilot study. In the sixth phase, auditing of 500 family folders is done to check content validity and construct validity of audit tool. Cronbach s alpha and principal component factor analysis method was applied to check the content validity and construct validity of the tool. Cronbach s alpha was calculated separately for :- (1) overall audit tool (2) sub groups of audit tool (3) individual items of the audit tool. Cronbach s alpha for (1) overall audit tool was 0.73, which shows that Tool was reliable and All the items of the tool were consistent and show unidimentionality, (As Cronbach s alpha 0.60 is reliable for any newly developed tool.) All the items of the tool were consistent and show unidimentionality, which means that they were measuring the specific domain for which they were constructed. When Cronbach s alpha was calculated for (2) sub groups of audit tool average value was 0.49 which shows that two groups, family background and maintenance of quality, have little contribution, while use of nursing process and child record have average contribution. The other two groups, family characteristics and maternal record, have maximum contribution for the overall construction of Audit Tool. When Cronbach s alpha was calculated for (3) the average value 169

5 of Cronbach s alpha if item deleted was The value of Cronbach s alpha for individual items of two groups, family background and Maintenance of quality was less than 0.31, so they were not retained. For the items of rest of the groups the value is equal to or more than 0.31, so they were retained. Table-1 : Cronbach's Alpha for Individual item in each Group of audit tool:- Subgroup No. of item Cronbach s Cronbach s alpha alpha if item deleted Family Background Maintenance of Quality Use of Nursing Process Family Characteristics Maternal record Child record To assess the adequacy as well as eligibility of tool items for undergoing factor analysis the Kmo value and p-value are calculated The Kmo value of data in this study was (The Kmo value of data should be 0.50.for proceeding for factor analysis.) The p-value of the data was 0.000; (the p-value should be less than 0.05 for proceeding for factor analysis.) So the data was eligible for factor analysis. Table-2 : Assessment of suitability of Audit Tool to carry out factor analysis Test Value KMO.575 Bartlett s test of sphericity chi-square p-value.000 Factor analysis of Audit Tool by Principal Component analysis after varimax rotation method indicate that total five factors should be retained in the Audit Tool on the basis of their factor loading effect. Factor 1 retains four items which have maximum factor loading effect in the construction of audit Tool. Factor 2 retains five items with factor loading effect less than factor 1. Factor 3 retains three items, for other two items signed by student and checked by teacher their variance was shared by the other factors (negative value). Factor 4 retains three items. Item identification has negative value Factor 5 retains three items. Factor analysis proves that on the basis of factor loading effect total eighteen items were retained in the tool.item such as identification, signed by student and checked by teacher should not be retained in the Tool because they were not measuring their construct exactly. 170

6 Table-3 : Factor analysis of Audit Tool by Principal Component Analysis after Varimax Rotation Components Factors Neonatal record Postnatal record Antenatal record Child under six record Intervention Diagnosis Assessment Dated Implementation Signed by student Record is Legible Date of starting the family health record Checked by teacher/supervisor Obstetrics record Women s record Identification Economic status Individual record tally with information Environment Social status Evaluation

7 Scree Plot The Scree plot is a graph of the eigenvalues against all the factors. The graph is useful for determining how many factors to be retained in the tool. The point of interest is where the curve starts to flatten. In this study from Scree plot, it is clear that first 2 factors had major contribution to the total variance (point of first inflection) and around five factors appear relevant in explaining data variance overall (point of second inflection).the first factor has eigen value 2.39, second factor mirror the almost same value 2.29, factor 3 has eigen value 1.60,factor four has eigen value 1.52,and factor five has eigen value 1.35.Subsequent to first five factors, the Scree plot curve is almost smoother without anymore inflection.so five factors and all the 22 items were retained on the basis of Scree plot test Eigenvalues Component Number Figure - 1 : Scree plot of eigenvalues against all factors 172

8 Discussion The Audit Tool is developed to audit the family health records. The Audit Tool is developed in six phases. This is compared with another study which was completed in six phases. 11 In first phase, the preliminary draft of Audit Tool was prepared.in second phase, content validity was checked with the help of Delphi technique. The Delphi is an iterative process designed to combine expert s opinion into group consesus 12 for the conventional Delphi a heterogeneous sample is used to ensure that the entire spectrum of opinion is determined 13. In present study heterogeneous panel of expert included the Nursing educators,senior residents and associate professors from School of Public health department.delphi technique employs a number of rounds in which questionnaire is sent out and are used until a consensus is reached Classical original Delphi used four rounds however this has been shortened to two or three rounds Current study has nine rounds, because the experts opinion was not reached into group consensus earlier as all the items of the questionnaire are finalized in ninth round. In third phase, necessar y modifications were done in the Tool. In fourth phase, feasibility of the Tool is assessed by doing a pilot study on family health records of village Dhanas. Pilot study help to identify wording difficulty as well as feasibility. 16 in fifth phase, final draft of the Tool was prepared. In sixth phase, auditing of 500 family health records of people of Dadu Majra Colony was done to check the content validity and construct validity of the Tool. Cronbach s alpha and principal component factor analysis method was applied to check the content validity and construct validity of the Tool. The Cronbach s alpha of newly developed tool should be 0.60 as per literature. The Cronbach s alpha of this study was 0.73.It means Audit tool is reliable. This was compared with Cronbach s alpha of other studies. In one study The Alcohol and Drug Confrontation Scale (ADCS) was developed. This scale had a 72-items.Cronbach s alpha of this scale was = In another study 15 items questionnaire was developed and tested on 69 subjects. The Cronbach s alpha was calculated directly on all items and on individual items of the tool. It was 0.89 & 0.82 respectively. 19. To check the eligibility of data for factor analysis minimum sample size should be total no of items multiplied by 10. In this study there are total 22 items. The sample size should be 220.As the sample size was 500 family health records.so the data was eligible for factor analysis. The data should have Kmo value 0.50 and p-value less than 0.05.for making the data eligible for factor analysis.the Kmo value of data in this study was and p-value was 0.000, which makes the data again eligible for factor analysis. This is compared with another study in which Kmo value was 0.52 and p-value was Principal component factor analysis method was used to select the items on the basis of factor loading effect. Total five factors with 18 items were retained Factor 1 retains 173

9 four items which have maximum value of first item (.90) Factor 2 retains five items. Factor 3 retains three items. For other two items signed by student and checked by teacher their variance is shared by the other factors (negative value). Factor 4 retains three items. Item identification has negative value Factor 5 retains three items. Factor analysis proves that on the basis of factor loading effect total eighteen items should be retained in the tool. Item such as identification, signed by student and checked by teacher should not be included in the Tool because they were not measuring their construct exactly. In one study where a short form of Chinese version of the State Anxiety Scale for children was developed, total five out of ten items were retained on the basis of factor loading effect 17. In another study 36 items out of 64 items were retained. 20. On the basis of Scree plot test five factors and 22 items were retained which was compared Scree plot test of other studies In the Alcohol and Drug Confrontation Scale (ADCS) 60 items were retained out of 72 original items. 18 In another study all 15 items were retained. 19 On the basis of expert s opinion, three items i.e. identification, signed by student and checked by teacher/supervisor should not be dropped as these are important. Because to measure the identification of a person, it is needed to gather information such as permanent residential proof, ration card/voter card/identity card. However, it is not feasible in this study as these details are not included in the family health record. The other items i.e. signed by the student and checked by teacher are there in the visit report of family health record, every record should be signed by the student who prepares it and every record should be checked by the teacher/ supervisor. The experts felt that these items were important and should be retained.so all the original items of the Audit Tool were retained. Contribution of all these items had already been proved & suppor ted with Cronbach s alpha and Scree plot test. This study has made contribution to evaluate the family health record maintained by the nursing students of NINE, PGIMER, Chandigarh with minimum time. Besides this, it will also help in improving the quality of documentation. This Tool enables the supervisors / faculty not to cross check the record by visiting the families personally as this Tool is self explanatory. Same study can be replicated with same audit Tool in a different setting i.e. in another college of nursing for auditing the health records maintained by their students. Audit Tool can be used by the faculty members to evaluate the student s records. References 1. Lancaster S. Community and Public health Nursing. 6th edition.mosby publication 2005 : Assaf A.F. AL Health care Quality: An International Prospective 2 nd Edition. Harcourt private limited 1995 : Kamal S Jogelkar. Hospital ward management; professional adjustment & trends in nursing, Mumbai; Vora medical publication 1990 :

10 4. Johnston G. Reviewing audit; barriers & facilitating factors for effective clinical audit. J Quality in health care 2000; 9: Bjovell C. Development of an audit instrument for nursing care plan in the patient record. J Quality in health care 2000; 9: Jamtvedt G. Audit & feedback: effects on professional practice & health care outcomes. J Quality in health care 2000; 7: Goel SL, R Kumar. Hospital administration & management.revised edition Deep & Deep publishers 2000 : Eilorat L, Cooper S. How to implement an audit to improve records. Nursing Times 1994; 90(35): Bailey Joan, Howse Ena. Resistance to documentation- a nursing research issue. International Journal of Nursing Studies1992; 4: V Binil.GV Nursing Audit. currentnursing.com //: Chris R. Record keeping, developing good practices nursing standards. September 18; 2002; 17: McKenna CM. The Delphi technique: a worthwhile approach for nursing. Journal of advanced nursing 1994; 19: Goodman CM. The Delphi technique: a critique journal of advanced nursing 1987: 12: Moore CM. Technique and the mail questionnaire. In: Group technique for idea building: applied social research methods. Series 9.Sage, Newbury Park 1987: Brer tta A. critical review of Delphi technique. Nurse Researcher 1996; 3(4): Green et al. Applying the Delphi Technique in a study of GP s information requirement. Health and social care in the community 1987; 5: Ho Cheung William. Violetta Lopez- Anxiety scale for children. International Journal of nursing studies 2007; 44: George A. Alcohol scale for alcoholic anonymous. Journal of studies on Alcohol 1996; 57: Mollema ed,heine R. Diabetes fear of injecting & self testing questionnaire. Diabetic care 2000;23: Joanne R.Dimension of caring Advances in Nursing science 2007;30: Acknowledgement Dr. Suksham Chopra is acknowledged for contributing in the literature review of the article Profile of nurses and their training needs published in volume 5(2) in April 2009, Page No