«SEPTEMBER 2015 This report was prepared by University Research Co., LLC (URC) for review by the United States Agency for International Development ...»
HAPIE Midline Assessment 13 Organizational Audit Information was sought mainly from the chief of the ward/unit and specialist responsible for medical care in the ward/unit. The area of assessment was the in-patient ward and other relevant units. For example, the delivery ward was checked for information of maternal cases while the cardiac intensive care unit was checked for all information relevant to treatment of those with AMI, and the emergency department and operating theatres were audited for all information pertinent to hip fracture.
Information was then verified through a review of records and observations. Similar to the hospital review, a maximum score was given when the information was verified with appropriate documentation and was consistent with findings in the service delivery units. In some cases, information needed to be validated against other relevant informants such as Medical Committee members and heads of department. Data were then entered into an excel sheet in the field.
Clinical Review The clinical review was based on the patient medical records. Data collection teams selected cases at random based on pre-determined diagnostic groups (see Table 7). At first, the team listed all cases with the diagnoses of interest during a period of one year preceding the data collection date. When the number of cases within that period was less than the sample size needed, cases were drawn from a longer period. Despite the small number of hip/femur fracture cases, we limited the data period up to 18 month before the study by looking at surgical registers. The number of hip fracture cases sampled was 30 in all but three hospitals: I (10 cases), H (29 cases) and E (16 cases). The variables of interest for each diagnostic group are listed in Table 8. Information extracted from patients’ clinical charts was checked for error and inconsistency.
Table 7. Clinical Conditions for Patient Record, Patient Experience
14 HAPIE Midline Assessment Diagnostic group Variables Routine monitoring Appropriate use of medications Acute Myocardialc Infarction Discharge status Adverse events Completeness of evaluation Routine monitoring Appropriate use of prophylaxis Hip and femur fracture Early mobilization Discharge status Adverse events Patient Interview Questionnaire A randomly selected sample of patients was given the questionnaire which contained questions about the patients’ experiences at the hospital for the present episode of care. Patients were selected from any one of the four departments: obstetrics/gynecology, pediatric, internal medicines and surgery. All were interviewed on the day of their discharge as close as possible to the time they were exiting the hospital. The data collection team intended randomly selecting patients for interview on one specific day during the data collection period. However, this was not feasible in all hospitals because the number of patient discharged per day was smaller than in the baseline due the JKN changes as noted below, especially in obstetric cases. In Hospitals B and D, the team extended the data collection to two days while in Hospitals D, E, and B, data collection was expanded to gynecology patients (within the same department as the obstetrics) because the number of patients discharged following childbirth was not adequate to fulfill the sample size requirements in two days. As with the clinical review data, completed questionnaires of patient interviews were checked for errors and inconsistencies during data entry.
Hospital-based Data Secondary data were collected from each hospital where possible on service quality, hospital company profile, hospital performance indicators, policies on regulations in all aspects, adverse events, and the formulation of committees to address deficiencies in patient care or hospital operations and, when available, the costs related to accreditation preparation.
Data were obtained from various unit such as the medical record unit, medical services unit, human resources unit and finance department. Data included annual reports, hospital performance indicators, financial statements, maternal death reports, clinical practice guideline, organizational structure, clinical pathways and adverse event reports. The degree of data completeness varied among hospitals.
Stakeholder Interview Informants interviewed during the midline study were key informants in accordance with criteria required by the qualitative study and most were the same as at baseline. Most were still in similar position as at the baseline. If informants from the baseline had moved to another position during midline, the team still attempted to conduct interview separately.
Interviews with stakeholders such as Chief of Medical Services, Hospital Accreditation Team, Finance Manager, Unit of hospital information system, and the Health Insurance Unit aimed to understand their opinions on the purpose of hospital accreditation, how the accreditation program was progressing from the perspectives of senior officials at participating hospitals, KARS, and MoH, and how the accreditation process influences the quality of care available delivered at the hospital in the HAPIE Midline Assessment 15 future. It also asked about changes in medical services and management since accreditation, hospital finances concerning accrediation, and changes in hospital policy regarding JKN.
Key informants included the Vice Director or the Head of Medical Services and the Head/Members of the accreditation teams. Using a snowball sampling technique, other informants were defined in the field. Table 9 list the informants involved in the study.
Table 9. Informants’ Characteristics
F. Data Analysis A profile of each hospital was generated to include basic parameters such the number of bed and surgical suites and the number of staff. It also included any specific details that the research assistants noted during their visit that were not recorded in other tools.
For the hospital review and organizational audit, once hospital visits were completed, the research team gathered to discuss the results to ensure standardization of scoring. It should be noted that a score of minimum compliance does not necessarily mean the standard is not in place at all. In some cases, information could not be verified with documentation although informants mentioned that the documentation was available, but not at the time that the data collector was there, because the person in charge was not in place or the supporting documents could not be located.
For the hospital review and organizational audit, the research team discussed the results obtained to ensure standardization of scoring. Hospital Review data were scored against standardized criteria using a 0-4 rating (0 is non-compliant, 4 is fully compliant). Organizational audit data were scored 0, 2 and 4. Data from clinical chart reviews and from patient experience questionnaires were analyzed by individual hospitals. Stakeholder interviews were recorded, transcribed and analyzed. Descriptive analysis was done to all variables. Chi-square statistical tests were used to examine associations between variables.
Descriptive analyses were done on all variables. Variables for which the unit of analysis was the patient (clinical review or survey), statistical test were applied. In the clinical review and patient interview, the dependent variable was the score of each variable while the independent variable was the time or period (baseline and midline). To determine the difference between the baseline and midline data infor clinical chart review and patient interview in each category of hospital, adjusting for covariates such as patient age and the type of insurance coverage reported.
Hospital Review Measurement of hospital compliance against the SANITAS criteria used the rating scale of 0-4. The scale represented the range from no or minimal compliance scoring 0 (e.g., in less than 5% of instances) to total or maximal compliance scoring 4 (e.g., compliance in more than 95% of instances).
A middle score of 2 represented compliance in a midway between these two extremes. If a hospital had a higher compliance than 5% but did not meet the criteria for the middle score, then a score of 1 16 HAPIE Midline Assessment was assigned; if compliance exceeded the middle score but was not fully compliant, then a score of 3 was assigned . These scores were simply compared from baseline to midline.
Organizational Audit Organizational audit data were given an ordinal score where 0 was for no or minimal compliance;
score 2 was for medium compliance and 4 was for total or maximal compliance.
Clinical Review Data from the clinical review were analyzed based on category that had been set, such as respondent characteristic, admissions, incompleteness of physical examination of pneumonia, completeness of laboratory examination for AMI patients, etc. The computed variable in each criterion are the critical elements performed in each diagnosis.
Patient Questionnaire Patient experience data were analyzed based on hospital group (JCI, KARS, NHA), as well as individual hospital in each group. Data on patient characteristics included the respondents’ highest education, age, sex, method of payment, and length of stay. Descriptive statistics were used for data on communication, satisfaction, hospitality, willingness to recommend the hospital to others, facilities worthiness, confidence in the clinical staff, and clarity of doctor’s discharge instruction.
Hospital-based Data Hospital-based indicators were analyzed between hospitals. This consisted of a hospital’s background or status (accreditation status, type, teaching hospital or not, number of patients, number of beds, etc.) and hospital performance indicators (BOR, LOS, TOI, NDR, emergency response time, hospitalacquired infections and other adverse events.) Stakeholder Interview Interview transcripts were entered directly into a matrix based on the themes of interest. Qualitative data were audio-recorded and interview transcripts were developed and grouped according to the themes studied.
G. Ethical Considerations Before agreeing to participate in the study, each hospital was informed of the data collection procedures for patients and staff including the time commitment, access to records, and feedback or sharing of results. Senior management and staff in relevant departments were formally briefed in advance on the nature and purpose of the study.
All study participants were assured that no individual identifying information would be recorded or used. As part of the informed consent process, all participants were be assured of their right to decline participation at any time during the interviews. All individuals, including managers, were assured that results from the study would be kept confidential and that only the research team would have access to raw data.
Patient interviews were conducted in separate rooms where staff and other patients could not overhear any part of the conversation. Participants were assured that their responses would not be communicated back to any hospital staff and that staff would not be informed of their participation.
H. Issues on Categorization of Studied Hospitals During development of this study, hospitals were chosen based on three hospital categories with three hospitals in each category.The classification was made based on their existing accreditation status and their plans between 2013 and 2016 determined from consultation with the MoH and hospital management. However, during this period, accreditation status and plans ofsome hospitals have changed such as Dr. Kariadi Hospital underwent their JCI mock survey in May 2014, while M.
Hoesin Hospital was planning their JCI mock survey on 2015. Persahabatan Hospital planned to undergo KARS accreditation in 2015.
HAPIE Midline Assessment 17 In 2012, Persahabatan Hospital was designated by the Ministry of Health to undergo JCI accreditation after undergoing KARS (2012) accreditation. However, Persahabatan was not ready for KARS (2012) accreditation because of deficiencies in facilities and infrastructure. Changes were also needed to meet patient safety and human resource management standards. M. Husein Hospital is also expected to seek JCI accreditation in 2015. Hospital management felt it could not yet meet all JCI standards but efforts were underway to improve systems and services to meet JCI standards. The hospital attempted KARS accreditation in 2014, and they succeeded in January, 2015. In Kariadi Hospital, according to one informant in management, they strive to meet both KARS and JCI accreditation using their own budget in 2015 without the mandate from the MoH.
III. RESULTS A. Clinical Charts and Patients Discharged The data collection aimed to perform clinical chart reviews on 30 casenotes for each diagnosis (vaginal delivery, pneumonia, acute myocardial infarct, and hip/collum femur fracture. The sample was selected randomly among admission in the 12 months prior to data collection, but for hip fracture diagnosis was expanded to one and half years because the cases were too few to meet the quota for 30 otherwise. When there were more than 30 admissions to choose from, the sample was selected randomly. Table 10 shows the number of charts reviewed in each category.
B. Hospital Review Out of the nine Class A hospitals, eight are main teaching hospitals and one is an affiliate teaching hospital. Every one has an area of focus (Center of Excellence) for medical service, such as trauma medicine, respiratory diseases, emergency medicine and cardiology. Table 11 lists key characeteristics of the nine hospitals.
Hospital Performance Indicators Performance indicators are data that reflect the extent to which an anticipated outcome is achieved or the quality of the processes leading to that outcome to which the standard has been set. Data
collected in the nine hospitals include: