Developing a Model to Predict High Health Care Utilization Among Patients in a New York City Safety Net System.

BACKGROUND: Health care facilities use predictive models to identify patients at risk of high future health care utilization who may benefit from tailored interventions. Previous predictive models that have focused solely on inpatient readmission risk, relied on commercial insurance claims data, or failed to incorporate social determinants of health may not be generalizable to safety net hospital populations. To address these limitations, we developed a payer-agnostic risk model for patients receiving care at the largest US safety net hospital system. METHODS: We transformed electronic health record and administrative data from 833,969 adult patients who received care during July 2016-July 2017 into demographic, utilization, diagnosis, medication, and social determinant variables (including homelessness and incarceration history) to predict health care utilization during the following year.We selected the final model by developing and validating multiple classification and regression models predicting 10+ acute days, 5+ acute days, or continuous acute days. We compared a portfolio of performance metrics while prioritizing positive predictive value for patients whose predicted utilization was among the top 1% to maximize clinical utility. RESULTS: The final model predicted continuous number of acute days and included 17 variables. For the top 1% of high acute care utilizers, the model had a positive predictive value of 47.6% and sensitivity of 17.3%. Previous health care utilization and psychosocial factors were the strongest predictors of future high acute care utilization. CONCLUSIONS: We demonstrated a feasible approach to predictive high acute care utilization in a safety net hospital using electronic health record data while incorporating social risk factors.

Differential Distortion of Purine Substrates by Human and Plasmodium falciparum Hypoxanthine-Guanine Phosphoribosyltransferase to Catalyse the Formation of Mononucleotides.

Plasmodium falciparum (Pf) hypoxanthine-guanine phosphoribosyltransferase (HGPRT) is a potential therapeutic target. Compared to structurally homologous human enzymes, it has expanded substrate specificity. In this study, 9-deazapurines are used as in situ probes of the active sites of human and Pf HGPRTs. Through the use of these probes it is found that non-covalent interactions stabilise the pre-transition state of the HGPRT-catalysed reaction. Vibrational spectra reveal that the bound substrates are extensively distorted, the carbonyl bond of nucleobase moiety is weakened and the substrate is destabilised along the reaction coordinate. Raman shifts of the human and Pf enzymes are used to quantify the differing degrees of hydrogen bonding in the homologues. A decreased Raman cross-section in enzyme-bound 9-deazaguanine (9DAG) shows that the phenylalanine residue (Phe186 in human and Phe197 in Pf) of HGPRT stacks with the nucleobase. Differential loss of the Raman cross-section suggests that the active site is more compact in human HGPRT as compared to the Pf enzyme, and is more so in the phosphoribosyl pyrophosphate (PRPP) complex 9DAG-PRPP-HGPRT than in 9-deazahypoxanthine (9DAH)-PRPP-HGPRT.

Solution structures of purine base analogues 6-chloroguanine, 8-azaguanine and allopurinol.

Analogues of purine bases are highly relevant in the biological context and have been implicated as drug molecules for therapy against a number of diseases. Additionally, these molecules have been implicated to have a role in the prebiotic RNA world. However, experimental data on the structures of these molecules in aqueous solution is lacking. In this work, we report the ultraviolet resonance Raman spectra of 6-chloroguanine, 8-azaguanine and allopurinol, obtained with 260 nm excitation. The reported spectra have been assigned to normal modes computed from density functional theory (B3LYP/6-31G (d,p)) calculations. This work has been useful in identifying the solution-state structures of these molecules at neutral pH. We find that the guanine analogues 6-chloroguanine and 8-azaguanine exist as keto-N9H and keto-N7H tautomers in solution, respectively. On the other hand, the hypoxanthine analogue allopurinol exists as a mixture of keto-N9H and keto-N8H tautomers in solution. We predict that this work would be particularly useful in future vibrational studies where these molecules are present in complexes with their target proteins.

Hypoxanthine guanine phosphoribosyltransferase distorts the purine ring of nucleotide substrates and perturbs the pKa of bound xanthosine monophosphate.

Enzymatic efficiency and structural discrimination of substrates from nonsubstrate analogues are attributed to the precise assembly of binding pockets. Many enzymes have the additional remarkable ability to recognize several substrates. These apparently paradoxical attributes are ascribed to the structural plasticity of proteins. A partially defined active site acquires complementarity upon encountering the substrate and completing the assembly. Human hypoxanthine guanine phosphoribosyltransferase (hHGPRT) catalyzes the phosphoribosylation of guanine and hypoxanthine, while the Plasmodium falciparum HGPRT (PfHGPRT) acts on xanthine as well. Reasons for the observed differences in substrate specificities of the two proteins are not clear. We used ultraviolet resonance Raman spectroscopy to study the complexes of HGPRT with products (IMP, GMP, and XMP), in both organisms, in resonance with the purine nucleobase electronic absorption. This led to selective enhancement of vibrations of the purine ring over those of the sugar-phosphate backbone and protein. Spectra of bound nucleotides show that HGPRT distorts the structure of the nucleotides. The distorted structure resembles that of the deprotonated nucleotide. We find that the two proteins assemble similar active sites for their common substrates. While hHGPRT does not bind XMP, PfHGPRT perturbs the pK(a) of bound XMP. The results were compared with the mutant form of hHGPRT that catalyzed xanthine but failed to perturb the pK(a) of XMP.

Structures, ionization equilibria, and tautomerism of 6-oxopurines in solution.

6-Oxopurine and its analogues form an important class of biological molecules that include nucleobases and their precursors and are substrates of a wide range of enzymes. Solution structures of purines have been debated in the literature because of the many possible tautomers and protonation states in which they can exist in solution. Substitutions on the pyrimidine and imidazole rings alter tautomerization and protonation equilibria, and as a consequence, the solution compositions and structures of closely related analogues can be significantly different. We have obtained resonance Raman spectra of 6-oxopurines: hypoxanthine, xanthine, their riboside phosphates, guanine monophosphate in the protonated and deprotonated forms with UV excitation at 260 nm. The species present in solution under different pH conditions were identified by isotopic labeling with deuterium as well as by comparison with extensive density functional theoretical calculations. At physiological pH, while N7H and N9H tautomeric forms of hypoxanthine exist in equilibrium, in xanthine, the additional carbonyl group at C2 shifts the equilibrium in favor of the N7H tautomer. The corresponding nucleotide of xanthine, xanthosine monophosphate, on the other hand, is in the anionic form (pK(a) 5.5). We find that Raman spectra show systematic shifts with change in the protonation state and substitution on the ring. In general, deprotonation of the neutral molecule is marked by a downshift in the observed Raman wavenumbers, and protonation is accompanied by an upshift.