Esrd v24 update

hccpy/_AGESEXV2.py

@@ -5,8 +5,11 @@ def get_ds(age, orec, medicaid, elig):
     if age < 65 and orec != "0":
         disabled = 1
     origds = 0
-    if orec in ("1", "3") and disabled == 0:
+    if orec in ("1") and disabled == 0:
         origds = 1
+    origesrd = 0
+    if orec in ("2", "3") and disabled == 0:
+        origesrd = 1
     if elig in {"NE", "SNPNE"}:
         if medicaid:
             elig += "_MCAID"
@@ -17,7 +20,7 @@ def get_ds(age, orec, medicaid, elig):
         else:
             elig += "_NORIGDIS_NE"
 
-    return disabled, origds, elig  
+    return disabled, origds, origesrd, elig
 
 
 

hccpy/_E2423T1M.py

@@ -0,0 +1,67 @@
+from collections import Counter
+
+def create_interactions(cc_lst, DISABL, age):
+
+    x = Counter(cc_lst) 
+    z = Counter()
+
+    # payable HCC's count interaction
+    hcccnt = len([k for k in x.keys() if "HCC" in k])
+    if hcccnt > 9:
+        x["D10P"] = 1
+    elif hcccnt > 0:
+        x["D{}".format(hcccnt)] = 1
+
+    # diagnostic categories
+    z["CANCER"] = max(x["HCC8"], x["HCC9"], x["HCC10"], 
+                        x["HCC11"], x["HCC12"])
+    z["DIABETES"] = max(x["HCC17"], x["HCC18"], x["HCC19"])
+    #z["IMMUNE"] = x["HCC47"]
+    z["CARD_RESP_FAIL"] = max(x["HCC82"], x["HCC83"], x["HCC84"])
+    z["CHF"] = x["HCC85"]
+    z["gCopdCF"] = max(x["HCC110"], x["HCC111"], x["HCC112"])
+    z["RENAL_V24"] = max(x["HCC134"], x["HCC135"], x["HCC136"], 
+                            x["HCC137"], x["HCC138"])
+    #z["COMPL"] = x["HCC176"]
+    z["SEPSIS"] = x["HCC2"]
+    z["gSubstanceUseDisorder_V24"] = max(x["HCC54"], x["HCC55"], x["HCC56"])
+    z["gPsychiatric_V24"] = max(x["HCC57"], x["HCC58"], x["HCC59"], x["HCC60"])
+
+    # community model interactions
+    x["HCC47_gCancer"] = x["HCC47"] * z["CANCER"]
+    x["DIABETES_CHF"] = z["DIABETES"] * z["CHF"]
+    x["CHF_gCopdCF"] = z["CHF"] * z["gCopdCF"]
+    x["HCC85_gRenal_V24"] =x["HCC85"] * z["RENAL_V24"]
+    x["gCopdCF_CARD_RESP_FAIL"] = z["gCopdCF"] * z["CARD_RESP_FAIL"]
+    x["HCC85_HCC96"] = x["HCC85"] * x["HCC96"]
+    x["gSubUseDs_gPsych_V24"] = z["gSubstanceUseDisorder_V24"] * z["gPsychiatric_V24"]
+
+    x["NONAGED_gSubUseDs_gPsych"] = DISABL * x["gSubUseDs_gPsych_V24"]
+    x["NONAGED_HCC6"] = DISABL * x["HCC6"]
+    x["NONAGED_HCC34"] = DISABL * x["HCC34"]
+    x["NONAGED_HCC46"] = DISABL * x["HCC46"]
+    x["NONAGED_HCC110"] = DISABL * x["HCC110"]
+    x["NONAGED_HCC176"] = DISABL * x["HCC176"]
+
+    # institutional model interactions
+    x["PRESSURE_ULCER"] = max(x["HCC157"], x["HCC158"], 
+                            x["HCC159"])
+    x["SEPSIS_PRESSURE_ULCER_V24"] = z["SEPSIS"] * x["PRESSURE_ULCER"]
+    x["SEPSIS_ARTIF_OPENINGS"] = z["SEPSIS"] * x["HCC188"]
+    x["ART_OPENINGS_PRESSURE_ULCER_V24"] = x["HCC188"] * x["PRESSURE_ULCER"]
+    x["gCopdCF_ASP_SPEC_B_PNEUM"] = z["gCopdCF"] * x["HCC114"]
+    x["ASP_SPEC_B_PNEUM_PRES_ULC_V24"] = x["HCC114"] * x["PRESSURE_ULCER"]
+    x["SEPSIS_ASP_SPEC_BACT_PNEUM"] = z["SEPSIS"] * x["HCC114"]
+    x["SCHIZOPHRENIA_gCopdCF"] = x["HCC57"] * z["gCopdCF"]
+    x["SCHIZOPHRENIA_CHF"] = x["HCC57"] * z["CHF"]
+    x["SCHIZOPHRENIA_SEIZURES"] = x["HCC57"] * x["HCC79"]
+
+    x["NONAGED_HCC85"] = DISABL * x["HCC85"]
+    x["NONAGED_PRESSURE_ULCER_V24"] = DISABL * x["PRESSURE_ULCER"]
+    x["NONAGED_HCC161"] = DISABL * x["HCC161"]
+    x["NONAGED_HCC39"] = DISABL * x["HCC39"]
+    x["NONAGED_HCC77"] = DISABL * x["HCC77"]
+
+    cc_lst = [k for k, v in x.items() if v > 0]
+
+    return cc_lst

hccpy/_E2423T1P.py

@@ -0,0 +1,169 @@
+def get_risk_dct(coefn, hcc_lst, age, sex, elig, origds, origesrd, full_partial_nondual, disabled, lti, graft_duration):
+
+    risk_dct = {}
+
+    # build demographic bracket strings and add to risk_dct
+
+    if elig in {"DI", "GI",}:
+        elig_sex = elig + "_" + sex
+    elif elig in {'GF','GNP'}:
+        # broken out by aged and non aged
+        aged = 'N' if disabled else 'A'
+        elig += aged
+        elig_sex = elig + "_" + sex
+    elif elig in {'DNE', 'GNE'}:
+        # new enrolle
+        elig_sex = "NE" + sex
+    elif elig in {"TRANSPLANT_KIDNEY_ONLY_1M",
+                  "TRANSPLANT_KIDNEY_ONLY_2M",
+                  "TRANSPLANT_KIDNEY_ONLY_3M"}:
+        elig_sex = elig
+
+
+
+    # build age bracket strings and add to risk_dct
+    # ESRD new enrolee age variables do not follow samme structure
+
+    if elig == 'GNE' and 65 <= age <= 69:
+        age_match = elig_sex + str(age)
+    elif elig in {"TRANSPLANT_KIDNEY_ONLY_1M",
+                  "TRANSPLANT_KIDNEY_ONLY_2M",
+                  "TRANSPLANT_KIDNEY_ONLY_3M"}:
+        age_match = elig_sex
+    else:
+        age_ranges = [x for x in coefn.keys() if elig_sex in x and elig in x and '_' in x[-5:]]
+        if not age_ranges:
+            print('\t',elig)
+            print('\t',disabled)
+            print('\t',elig_sex)
+            print('\t',age_ranges)
+        age_match = ""
+        for age_range in age_ranges:
+
+            age_tokens = age_range[-5:].replace(sex, "").split("_") 
+            lb, ub = 0, 999 
+            if len(age_tokens) == 1:
+                lb = int(age_tokens[0])
+                ub = lb + 1
+            elif age_tokens[1] == "GT":
+                lb = int(age_tokens[0])
+            else: 
+                lb = int(age_tokens[0])
+                ub = int(age_tokens[1]) + 1
+            if lb <= age < ub:
+                if elig in {'DNE', 'GNE'}:
+                    age_match = elig_sex + age_range.split(elig_sex)[1]
+                else:
+                    age_match = age_range
+                break
+
+    # Dual status interactions with Age and Sex used in dialysis CE model;
+    elig_fbpb = ''
+    if elig == "DI" and full_partial_nondual != 'N':
+        if full_partial_nondual == 'F':
+            elig_fbpb = elig + '_FBDual'
+        else:
+            elig_fbpb = elig + '_PBDual'
+        if sex == 'M':
+            elig_fbpb += '_Male'
+        else:
+            elig_fbpb += '_Female'
+        if disabled:
+            elig_fbpb += '_NonAged'
+        else:
+            elig_fbpb += '_Aged'
+
+    # Originally Disabled Interactions with Sex used in dialysis CE model and functioning graft community aged models;
+    elig_origds = ''
+    if origds > 0:
+        elig_origds = elig + "_OriginallyDisabled_"
+        if sex == "M":
+            elig_origds += "Male"
+        else:
+            elig_origds += "Female"
+
+    # Originally ESRD interactions with Sex used in dialysis CE model;
+    elig_origesrd = ''
+    if elig == 'DI' and origesrd > 0:
+        elig_origesrd = elig + "_Originally_ESRD_"
+        if sex == "M":
+            elig_origesrd += "Male"
+        else:
+            elig_origesrd += "Female"
+
+    # LTI interactions with Aged used in dialysis CE model;
+    elig_ltia = ''
+    if elig == "DI" and lti > 0:
+        elig_ltia = elig + '_LTI_'
+        if disabled:
+            elig_ltia += 'NonAged'
+        else:
+            elig_ltia += 'Aged'
+
+    # dualstatus (ND_PBD / FBD), origds (NORIGDIS / ORIGDIS), modelcode ('', G), sex_age_band (== age_match)
+
+    base_demo_variable = ''
+    if elig in {'DNE', 'GNE'}:
+        base_demo_variable = '_'.join([
+            elig,
+            'FBD' if full_partial_nondual == 'F' else 'ND_PBD',
+            'ORIGDIS' if origds > 0 else 'NORIGDIS',
+            age_match if elig == 'DNE' else 'G_' + age_match
+        ])
+    else:
+        base_demo_variable = age_match
+
+    fgf_variable = '' # functioning graft factor transplant bump-up factors
+    pbd_factor_variable = '' # partial benefit dual status bump up factors
+    actadj_variable = ''
+    if 'G' in elig:
+        if graft_duration:
+            fgf_variable = '_'.join([
+                'FGI' if lti > 0 else 'FGC',
+                'LT65' if disabled > 0 else 'GE65',
+                graft_duration,
+                'FBD' if full_partial_nondual == 'F' else 'ND_PBD',
+            ])
+        if full_partial_nondual == 'P' and not 'NE' in elig:
+            pbd_factor_variable = '_'.join([
+                'FGI' if lti > 0 else 'FGC',
+                'PBD',
+                'LT65' if disabled > 0 else 'GE65',
+                'flag'
+            ])
+
+        # Actuarial adjustment factor for new enrollees functioning graft model transplant bumps
+        if elig == 'GNE' and graft_duration:
+            actadj_variable = 'ActAdj_'
+            actadj_variable += graft_duration
+
+    actadj_factor = coefn.get(actadj_variable, 1.0)
+
+    if base_demo_variable:
+        risk_dct[base_demo_variable] = coefn.get(base_demo_variable, 0.0)
+
+    if elig_fbpb:
+        risk_dct[elig_fbpb] = coefn.get(elig_fbpb, 0.0)
+
+    if elig_origds:
+        risk_dct[elig_origds] = coefn.get(elig_origds, 0.0)
+
+    if elig_origds:
+        risk_dct[elig_origds] = coefn.get(elig_origds, 0.0)
+
+    if elig_ltia:
+        risk_dct[elig_ltia] = coefn.get(elig_ltia, 0.0)
+
+    if fgf_variable:
+        risk_dct[fgf_variable] = coefn.get(fgf_variable, 0.0)
+
+    if pbd_factor_variable:
+        risk_dct[pbd_factor_variable] = coefn.get(pbd_factor_variable, 0.0)
+
+
+    # # build hcc factor strings and add to risk_dict
+    for hcc in hcc_lst:
+        elig_hcc = elig + '_' + hcc
+        risk_dct[elig_hcc] = coefn.get(elig_hcc, 0.0)
+
+    return risk_dct, actadj_factor