PENGEMBANGAN MODEL PREDIKTIF DALAM MENGUKUR KINERJA PERAWAT DAN KEBIJAKANNYA DI RUMAH SAKIT UMUM DAERAH TANGERANG (2011)

HM. HAFIZURRACHMAN, Prof. dr. Laksono Trisnantoro, M.Sc, Ph.D

2012 | Disertasi | S3 Kedokteran Umum

Latar belakang: Di rumah sakit Kabupaten Tangerang, perawat adalah petugas kesehatan utama yang melayani pasien di rumah sakit dan harus memiliki kinerja yang maksimal sehingga pasien akan mendapatkan pelayanan prima. Untuk ini perlu diukur kinerja mereka secara teratur dengan alat yang handal, mudah dan prediktif melalui pengembangan model prediktif dan dilengkapi dengan penyusunan kebijakan khusus. Tujuan: Mengkaji dan mengembangkan model prediksi pengukuran kinerja perawat rumah sakit serta Implikasi kebijakannya di RSUD Tangerang. Metode: Desain penelitian Mixed Method. Data diolah dengan Smart PLS membentuk Structural Equation Model untuk membuktikan hubungan confirmatory dan exploratory pada teorema 8 variabel laten yang memiliki 4-6 indikator yang mempengaruhi variabel kinerja perawat serta mencari besaran hubungan antarvariabel tersebut (nilai rho), sehingga terbentuk persamaan liner model. Hasilnya untuk membuat program software terhadap pengukuran kinerja prediksi perawat. Sampelnya 250 perawat dengan 163 perawat bertindak sebagai rater. Sampel diambil secara multistage random sampling. Kemudian dilanjutkan dengan penelitian kualitatif kebijakan terhadap terjadinya hal di atas. Informan 29 perawat, berasal dari dalam dan luar RSUD Tangerang. Proses pengambilan informasi melalui telaah dokumen, obervasi lapangan, focus group discussion dan wawancara mendalam. Triangulasi informasi dilaksanakan untuk mendapatkan hasil yang terpercaya dan akurat. Pendekatan Policy Analytical Process menggunakan kriteria Boulton diterapkan sehingga diketahui permasalahan kebijakan tentang keperawatan umum dan aturan kinerja keperawatan di RSUD tangerang yang terkait dengan fit to work serta melindungi dari risiko dan bahaya pekerjaan keperawatan dan disajikan melalui format analisis tentang kebijakan (analysis of policy). Selanjutnya dengan menggabungkan ke dua temuan dari ke dua desain penelitian tersebut akan menghasilkan 3 strategi untuk dipilih satu strategi saja dan pilihan jatuh pada strategi C. Strategi ini dijabarkan kedalam suatu rancangan kebijakan keperawatan berbasis kinerja prediktif di RSUD Tangerang yang dituangkan atau ditulis kedalam format analisis untuk kebijakannya (analysis for policy). Penelitian dilakukan 7 bulan mulai Febuari sampai dengan Agustus 2011. Hasil: Di RSUD Tangerang hasil perhitungan dengan Smart-PLS didapatkan pada 35 hubungan antarvariabel yang dibentuk pada model struktur awal, hanya menghasilkan 18 hubungan antarvariabel dengan nilai goodness of fit yang signifikan dan pada model modifikasi struktur akhir meningkat menjadi 19 hubungan antarvariabel yang signifikan. Pada model struktur akhir tersebut membentuk 3 variabel yang memiliki pengaruh secara langsung dengan goodness of fit yang signifikan terhadap variabel kinerja prediktif yaitu variabel lingkungan kehidupan, ability dan sejarah kesehatan keluarga perawat. Adapun pengaruh semua variabel prediktor terhadap variabel kinerja perawat pada model struktur akhir 89,24% dengan nilai Q-predictive relevance 93,88%; yang mana hasilnya digunakan untuk membuat software model kinerja prediktif perawat. Sebagai lanjutan penelitan kualitatif kebijakan keperawatan menghasilkan strategi C berupa rumusan rancangan kebijakan keperawatan berbasis kinerja prediktif. Kesimpulan dan Saran: Variabel kinerja prediksi perawat RSUD Tangerang dipengaruhi oleh berbagai variabel diantaranya ke 8 variabel prediktor kinerja sebagai sumber penyusunan strategi kebijakan keperawatan berbasis kinerja yang mengutamakan kesesuaian penempatan kerja, dan melindungi perawat dari risiko dan bahaya pekerjaan keperawatan, sehingga managemen RS dapat melakukan tindakan antisipatif bila ditemukan penurunan kinerja perawat dan kebijakan yang dibuat menjamin para perawat bekerja dengan maksimal

Background: In District hospital of Tangerang, nurses were major health workers who serve patients in hospitals and should have a maximum performance so that patients would get prime services. For this needs to be measured their performance regularly with a reliable tool, easy and predictive through the development of predictive models and comes with the preparation of specific policies. Objectives: Assess and develop predictive models of nurse hospital performance measurement as well as his policy implications at the hospital of Tangerang. Methods: Research design was mixed method. Data was processed with Smart-PLS formed a Structural Equation Model to prove the confirmatory and exploratory relationships in theorem 8 latent variables that have a 4-6 indicator variables that affect the nurse performance variable and to find the magnitude relationship among variables (rho values) so that forming an equation of linear model. The result was to create a software program to measure the predicted of performance of nurses. Sample 250 nurses and the others 163 nurses acted as raters. The way samples were taken with multistage random sampling. Then proceed with the policy qualitative research to the occurrence of it the above. 29 nurses as an informants, comes from within and outside the Tangerang Hospital. Information collected by the way of document review, a field observation, focus group discussions and depth interviews. Triangulation of information carried out to obtain reliable and accurate results. The Approach \\\"Policy Process Analysis\\\" using the Boulton criteria, so that known nursing policy issues concerning general nursing policy and nurse's performance rules in Hospital of Tangerang associated with suitability to worked place and protect nurses from risks and dangers of work of nursing and presented via the format of policy analysis (analysis of Policy). Furthermore, by combining the two findings from the two research designs produce three strategies to be chosen one strategy and the choice fell on a strategy of C. This strategy was translated into a draft of a nursing policy based on a predictive of nurse's performance at hospital of Tangerang that was poured or written into a format of \\\"analysis for policy (analysis for policy). The study was conducted 7 months, from February to August 2011. Results: In District hospital of Tangerang, calculations with Smart-PLS obtained at 35 relationships among the variables at the initial structure model, produces 18 relationships between the variables with a value of goodness of fit was significant and at the model final modification of structure increased to 19 significant relationships among the variables. At the final model for the three variables have direct influence with the \\\"goodness of fit\\\" significantly to the predictive performance variable. The variables were life environment variable, abilities and history of nurse's family health. The influence of all predictor variables on the performance variables nurse at the final model was 89.24% with the Q-predictive value relevance was 93.88%. The results were used for the software development of predictive model of nurses’ performance. As a continuation, a qualitative research about nursing policy generates a strategy C which was a draft of nursing based on a nurse’s predictive' performance in Tangerang District Hospital. Conclusion and Suggestions: Variables predictive of performance Tangerang District Hospital nurses were affected by many variables including the eight predictor of performance variables as a source for formulating of nursing policy strategy, based on performance which was prioritize the suitability of work placements and protects nurses from risk and hazard of work of nursing, so hospital management can take anticipative action if found to decrease performance of nurses and policies were made to ensure the nurses work with the maximum.

Kata Kunci : kinerja perawat - structure equation model - policy analysis - regulasi keperawatan.