diff --git a/pr_cvd_female_raw_bmirep_qrisk2.do b/pr_cvd_female_raw_bmirep_qrisk2.do
new file mode 100644
index 0000000..cfe0147
--- /dev/null
+++ b/pr_cvd_female_raw_bmirep_qrisk2.do
@@ -0,0 +1,172 @@
+*use a dataset including 
+cap prog drop cvd_female_raw_qrisk2
+program define cvd_female_raw_qrisk2
+
+noi di "Variables you need:"
+noi di "age -- age as integer"
+noi di	"b_AF -- atrial fibrilation as binary 01"
+noi di "b_ra -- rheumatoid arthritis as binary 01"
+noi di "b_renal -- rename disease as binary 01"
+noi di "b_treatedhyp -- hypertension treatment as binary 01"
+noi di "b_type1 -- type 1 diabetes as binary 01"
+noi di "b_type2 -- type 2 diabetes as binary 01"
+noi di "bmi -- bmi as continuous numeric, replaced with 20 if below 20"
+noi di "ethrisk -- ethnicity as categorical integer (0-9)"
+noi di "fh_cvd -- family history as binary 01"
+noi di "rati -- HDL LDL cholesterol ratio as continuous numeric"
+noi di "sbp -- systolic blood pressure as continuous numeric"
+noi di "smoke_cat -- smoking as integer category (0-4)"
+noi di "sex  -- sex as 1=male 2=female"
+
+
+
+cap gen score=.
+gen double rec_num = _n
+preserve
+
+keep if sex==2
+*replace bmi=20 if bmi<20
+
+
+local surv=10
+forvalues i = 0/15 {
+	gen survivor`i'=0
+}
+replace survivor10=0.989747583866119
+*this part is added in incase they want to do risk scores over a different number of years but they haven't actually done the groundwork yet.
+
+gen Iethrisk0 = 0
+gen Iethrisk1 = 0 /*5*/
+gen Iethrisk2 = 0.2574099349831925900000000 /*3*/
+gen Iethrisk3 = 0.6129795430571779400000000 /*1*/
+gen Iethrisk4 = 0.3362159841669621300000000 /*2*/
+gen Iethrisk5 = 0.1512517303224336400000000 /*4*/
+gen Iethrisk6 = -0.1794156259657768100000000 /*7*/
+gen Iethrisk7 = -0.3503423610057745400000000 /*9*/
+gen Iethrisk8 = -0.2778372483233216800000000 /*8*/
+gen Iethrisk9 = -0.1592734122665366000000000 /*6*/
+
+gen Ismoke0=0
+gen Ismoke1=0.2119377108760385200000000
+gen Ismoke2=0.6618634379685941500000000
+gen Ismoke3=0.7570714587132305600000000
+gen Ismoke4=0.9496298251457036000000000
+
+gen double dage = age
+replace dage=dage/10
+gen double age_2 = dage
+gen double age_1 = dage^0.5
+gen double dbmi = bmi
+replace dbmi=dbmi/10
+gen double bmi_2 = ln(dbmi)*dbmi^(-2)
+gen double bmi_1 = dbmi^(-2)
+
+replace age_1 = age_1 - 2.086397409439087
+replace age_2 = age_2 - 4.353054523468018
+replace bmi_1 = bmi_1 - 0.152244374155998
+replace bmi_2 = bmi_2 - 0.143282383680344
+replace rati = rati - 3.506655454635620
+replace sbp = sbp - 125.040039062500000
+replace town = town - 0.416743695735931
+
+gen double a=0
+
+
+
+
+
+
+/* The conditional sums */
+forvalues i=1/9 {
+replace a = a+ Iethrisk`i' if ethrisk==`i'
+} 
+forvalues i=0/4{
+replace a = a+ Ismoke`i' if smoke_cat==`i' 
+}
+
+	/* Sum from continuous values */
+
+
+ replace a = a+ age_1 * 4.4417863976316578000000000 
+ replace a = a+ age_2 * 0.0281637210672999180000000 
+ replace a = a+ bmi_1 * 0.8942365304710663300000000 
+ replace a = a+ bmi_2 * -6.5748047596104335000000000 
+ replace a = a+ rati * 0.1433900561621420900000000 
+ replace a = a+ sbp * 0.0128971795843613720000000 
+ replace a = a+ town * 0.0664772630011438850000000 
+
+
+
+	/* Sum from boolean values */
+
+ replace a = a+ b_AF * 1.6284780236484424000000000 
+ replace a = a+ b_ra * 0.2901233104088770700000000 
+ replace a = a+ b_renal * 1.0043796680368302000000000 
+ replace a = a+ b_treatedhyp * 0.6180430562788129500000000 
+ replace a = a+ b_type1 * 1.8400348250874599000000000 
+ replace a = a+ b_type2 * 1.1711626412196512000000000 
+ replace a = a+ fh_cvd * 0.5147261203665195500000000 
+
+	/* Sum from interaction terms */
+
+ replace a = a+ age_1 * 1 * 0.7464406144391666500000000 if smoke_cat==1 
+ replace a = a+ age_1 * 1 * 0.2568541711879666600000000 if smoke_cat==2
+ replace a = a+ age_1 * 1 * -1.5452226707866523000000000 if smoke_cat==3
+ replace a = a+ age_1 * 1 * -1.7113013709043405000000000 if smoke_cat==4
+ replace a = a+ age_1 * b_AF * -7.0177986441269269000000000 
+ replace a = a+ age_1 * b_renal * -2.9684019256454390000000000 
+ replace a = a+ age_1 * b_treatedhyp * -4.2219906452967848000000000 
+ replace a = a+ age_1 * b_type1 * 1.6835769546040080000000000 
+ replace a = a+ age_1 * b_type2 * -2.9371798540034648000000000 
+ replace a = a+ age_1 * bmi_1 * 0.1797196207044682300000000 
+ replace a = a+ age_1 * bmi_2 * 40.2428166760658140000000000 
+ replace a = a+ age_1 * fh_cvd * 0.1439979240753906700000000 
+ replace a = a+ age_1 * sbp * -0.0362575233899774460000000 
+ replace a = a+ age_1 * town * 0.3735138031433442600000000 
+ replace a = a+ age_2 * 1 * -0.1927057741748231000000000 if smoke_cat==1
+ replace a = a+ age_2 * 1 * -0.1526965063458932700000000 if smoke_cat==2
+ replace a = a+ age_2 * 1 * 0.2313563976521429400000000 if smoke_cat==3
+ replace a = a+ age_2 * 1 * 0.2307165013868296700000000 if smoke_cat==4
+ replace a = a+ age_2 * b_AF * 1.1395776028337732000000000 
+ replace a = a+ age_2 * b_renal * 0.4356963208330940600000000 
+ replace a = a+ age_2 * b_treatedhyp * 0.7265947108887239600000000 
+ replace a = a+ age_2 * b_type1 * -0.6320977766275653900000000 
+ replace a = a+ age_2 * b_type2 * 0.4023270434871086800000000 
+ replace a = a+ age_2 * bmi_1 * 0.1319276622711877700000000 
+ replace a = a+ age_2 * bmi_2 * -7.3211322435546409000000000 
+ replace a = a+ age_2 * fh_cvd * -0.1330260018273720400000000 
+ replace a = a+ age_2 * sbp * 0.0045842850495397955000000 
+ replace a = a+ age_2 * town * -0.0952370300845990780000000 
+
+ 
+ gen double scoref = exp(a)
+ replace scoref=survivor`surv'^scoref
+ replace  scoref = 1-scoref
+ replace  scoref=100*scoref
+ *gen double scoref = 100.0 * (1 - survivor`surv'^(exp(a)))
+ 
+keep scoref patid rec_num
+
+tempfile temper
+save `temper' , replace
+restore
+merge 1:1 patid rec_num using `temper' , nogen
+drop rec_num
+
+replace score=scoref if scoref!=.
+drop scoref
+ 
+end
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
diff --git a/pr_cvd_female_raw_bmirep_qrisk3.do b/pr_cvd_female_raw_bmirep_qrisk3.do
new file mode 100644
index 0000000..9ae7daf
--- /dev/null
+++ b/pr_cvd_female_raw_bmirep_qrisk3.do
@@ -0,0 +1,188 @@
+*use a dataset including 
+cap prog drop cvd_female_raw_qrisk3
+program define cvd_female_raw_qrisk3
+
+noi di "Variables you need:"
+noi di "age -- age as integer"
+noi di	"b_AF -- atrial fibrilation as binary 01"
+noi di "b_ra -- rheumatoid arthritis as binary 01"
+noi di "b_renal -- rename disease as binary 01"
+noi di "b_treatedhyp -- hypertension treatment as binary 01"
+noi di "b_type1 -- type 1 diabetes as binary 01"
+noi di "b_type2 -- type 2 diabetes as binary 01"
+noi di "bmi -- bmi as continuous numeric, replaced with 20 if below 20"
+noi di "ethrisk -- ethnicity as categorical integer (0-9)"
+noi di "fh_cvd -- family history as binary 01"
+noi di "rati -- HDL LDL cholesterol ratio as continuous numeric"
+noi di "sbp -- systolic blood pressure as continuous numeric"
+noi di "smoke_cat -- smoking as integer category (0-4)"
+noi di "sex  -- sex as 1=male 2=female"
+
+noi di "b_atypicalantipsy -- atypical antipsychotic prescribing as binary 01"
+noi di "b_semi - severe mental illness as binary 01"
+noi di "sbp_s5 - systolic blood pressure variability as continuous numeric"
+noi di "b_impotence2 - erectile dysfunction as binary 01"
+noi di "b_sle - systemic lupus erythematosus as binary 01"
+noi di "b_corticosteroids - corticosteroid prescribing as binary 01"
+
+
+cap gen score=.
+gen double rec_num = _n
+preserve
+
+keep if sex==2
+*replace bmi=20 if bmi<20
+
+
+local surv=10
+forvalues i = 0/15 { 
+	gen survivor`i'=0
+}
+replace survivor10=0.988876402378082
+*this part is added in incase they want to do risk scores over a different number of years but they haven't actually done the groundwork yet.
+
+gen Iethrisk0 = 0
+gen Iethrisk1 = 0
+gen Iethrisk2 = 0.2804031433299542500000000
+gen Iethrisk3 = 0.5629899414207539800000000
+gen Iethrisk4 = 0.2959000085111651600000000
+gen Iethrisk5 = 0.0727853798779825450000000
+gen Iethrisk6 = -0.1707213550885731700000000
+gen Iethrisk7 = -0.3937104331487497100000000
+gen Iethrisk8 = -0.3263249528353027200000000 
+gen Iethrisk9 = -0.1712705688324178400000000
+
+gen Ismoke0=0
+gen Ismoke1=0.1338683378654626200000000
+gen Ismoke2=0.5620085801243853700000000
+gen Ismoke3=0.6674959337750254700000000
+gen Ismoke4=0.8494817764483084700000000
+
+gen double dage = age
+replace dage=dage/10
+gen double age_1 = dage^(-2)
+gen double age_2 = dage
+gen double dbmi = bmi
+replace dbmi=dbmi/10
+gen double bmi_1 = dbmi^(-2)
+gen double bmi_2 = ln(dbmi)*dbmi^(-2)
+
+replace age_1 = age_1 - 0.053274843841791
+replace age_2 = age_2 - 4.332503318786621
+replace bmi_1 = bmi_1 - 0.154946178197861
+replace bmi_2 = bmi_2 - 0.144462317228317
+replace rati = rati - 3.476326465606690
+replace sbp = sbp - 123.130012512207030
+replace sbps5 = sbps5 - 9.002537727355957
+replace town = town - 0.392308831214905
+
+/* Start of Sum */
+gen double a=0
+
+
+/* The conditional sums */
+forvalues i=1/9 {
+replace a = a+ Iethrisk`i' if ethrisk==`i'
+} 
+forvalues i=0/4{
+replace a = a+ Ismoke`i' if smoke_cat==`i' 
+}
+
+	/* Sum from continuous values */
+
+
+ replace a = a+ age_1 * -8.1388109247726188000000000 
+ replace a = a+ age_2 * 0.7973337668969909800000000 
+ replace a = a+ bmi_1 * 0.2923609227546005200000000 
+ replace a = a+ bmi_2 * -4.1513300213837665000000000 
+ replace a = a+ rati * 0.1533803582080255400000000 
+ replace a = a+ sbp * 0.0131314884071034240000000
+ replace a = a+ sbps5 * 0.0078894541014586095000000 if sbps5 != .
+ replace a = a+ town * 0.0772237905885901080000000 
+
+
+
+	/* Sum from boolean values */
+
+ replace a = a+ b_AF * 1.5923354969269663000000000
+ replace a = a+ b_atypicalantipsy * 0.2523764207011555700000000
+ replace a = a+ b_corticosteroids * 0.5952072530460185100000000
+ replace a = a+ b_migraine * 0.3012672608703450000000000
+ replace a = a+ b_ra * 0.2136480343518194200000000 
+ replace a = a+ b_renal * 0.6519456949384583300000000
+ replace a = a+ b_semi * 0.1255530805882017800000000
+ replace a = a+ b_sle * 0.7588093865426769300000000
+ replace a = a+ b_treatedhyp * 0.5093159368342300400000000 
+ replace a = a+ b_type1 * 1.7267977510537347000000000 
+ replace a = a+ b_type2 * 1.0688773244615468000000000 
+ replace a = a+ fh_cvd * 0.4544531902089621300000000 
+
+	/* Sum from interaction terms */
+
+ replace a = a+ age_1 * -4.7057161785851891000000000 if smoke_cat==1 
+ replace a = a+ age_1 * -2.7430383403573337000000000 if smoke_cat==2
+ replace a = a+ age_1 * -0.8660808882939218200000000 if smoke_cat==3
+ replace a = a+ age_1 * 0.9024156236971064800000000 if smoke_cat==4
+ replace a = a+ age_1 * b_AF * 19.9380348895465610000000000
+ replace a = a+ age_1 * b_corticosteroids * -0.9840804523593628100000000
+ replace a = a+ age_1 * b_migraine * 1.7634979587872999000000000
+ replace a = a+ age_1 * b_renal * -3.5874047731694114000000000
+ replace a = a+ age_1 * b_sle * 19.6903037386382920000000000
+ replace a = a+ age_1 * b_treatedhyp * 11.8728097339218120000000000 
+ replace a = a+ age_1 * b_type1 * -1.2444332714320747000000000 
+ replace a = a+ age_1 * b_type2 * 6.8652342000009599000000000 
+ replace a = a+ age_1 * bmi_1 * 23.8026234121417420000000000 
+ replace a = a+ age_1 * bmi_2 * -71.1849476920870070000000000 
+ replace a = a+ age_1 * fh_cvd * 0.9946780794043512700000000 
+ replace a = a+ age_1 * sbp * 0.0341318423386154850000000 
+ replace a = a+ age_1 * town * -1.0301180802035639000000000 
+ replace a = a+ age_2 * -0.0755892446431930260000000 if smoke_cat==1
+ replace a = a+ age_2 * -0.1195119287486707400000000 if smoke_cat==2
+ replace a = a+ age_2 * -0.1036630639757192300000000 if smoke_cat==3
+ replace a = a+ age_2 * -0.1399185359171838900000000 if smoke_cat==4
+ replace a = a+ age_2 * b_AF * -0.0761826510111625050000000
+ replace a = a+ age_2 * b_corticosteroids * -0.1200536494674247200000000
+ replace a = a+ age_2 * b_migraine * -0.0655869178986998590000000
+ replace a = a+ age_2 * b_renal * -0.2268887308644250700000000
+ replace a = a+ age_2 * b_sle * 0.0773479496790162730000000
+ replace a = a+ age_2 * b_treatedhyp * 0.0009685782358817443600000 
+ replace a = a+ age_2 * b_type1 * -0.2872406462448894900000000 
+ replace a = a+ age_2 * b_type2 * -0.0971122525906954890000000 
+ replace a = a+ age_2 * bmi_1 * 0.5236995893366442900000000 
+ replace a = a+ age_2 * bmi_2 * 0.0457441901223237590000000 
+ replace a = a+ age_2 * fh_cvd * -0.0768850516984230380000000 
+ replace a = a+ age_2 * sbp * -0.0015082501423272358000000 
+ replace a = a+ age_2 * town * -0.0315934146749623290000000 
+
+ 
+ gen double scoref = exp(a)
+ replace scoref=survivor`surv'^scoref
+ replace  scoref = 1-scoref
+ replace  scoref=100*scoref
+ *gen double scoref = 100.0 * (1 - survivor`surv'^(exp(a)))
+ 
+keep scoref patid rec_num
+
+tempfile temper
+save `temper' , replace
+restore
+merge 1:1 patid rec_num using `temper' , nogen
+drop rec_num
+
+replace score=scoref if scoref!=.
+drop scoref
+ 
+end
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
diff --git a/pr_cvd_male_raw_bmirep_qrisk2.do b/pr_cvd_male_raw_bmirep_qrisk2.do
new file mode 100644
index 0000000..f115033
--- /dev/null
+++ b/pr_cvd_male_raw_bmirep_qrisk2.do
@@ -0,0 +1,175 @@
+*use a dataset including 
+cap prog drop cvd_male_raw_qrisk2
+program define cvd_male_raw_qrisk2
+
+noi di "Variables you need:"
+noi di "age -- age as integer"
+noi di	"b_AF -- atrial fibrilation as binary 01"
+noi di "b_ra -- rheumatoid arthritis as binary 01"
+noi di "b_renal -- rename disease as binary 01"
+noi di "b_treatedhyp -- hypertension treatment as binary 01"
+noi di "b_type1 -- type 1 diabetes as binary 01"
+noi di "b_type2 -- type 2 diabetes as binary 01"
+noi di "bmi -- bmi as continuous numeric"
+noi di "ethrisk -- ethnicity as categorical integer (0-9)"
+noi di "fh_cvd -- family history as binary 01"
+noi di "rati -- HDL LDL cholesterol ratio as continuous numeric"
+noi di "sbp -- systolic blood pressure as continuous numeric"
+noi di "smoke_cat -- smoking as integer category (0-4)"
+noi di "town -- townsend deprivation score as continuous numeric -- not clear how this is provided, as score or quintile"
+noi di "sex  -- sex as 1=male 2=female"
+
+cap gen score=.
+gen double rec_num=_n
+preserve
+keep if sex==1
+*replace bmi=20 if bmi<20
+
+
+local surv=10
+forvalues i = 0/15 {
+	gen survivor`i'=0
+}
+replace survivor10=0.978794217109680
+*this part is added in incase they want to do risk scores over a different number of years but they haven't actually done the groundwork yet.
+
+gen Iethrisk0 = 0
+gen Iethrisk1 = 0 /*5*/
+gen Iethrisk2 = 0.3173321430481919100000000 /*3*/
+gen Iethrisk3 = 0.4738590786081115500000000 /*2*/
+gen Iethrisk4 = 0.5171314655968145500000000 /*1*/
+gen Iethrisk5 = 0.1370301157366419200000000 /*4*/
+gen Iethrisk6 = -0.3885522304972663900000000 /*8*/
+gen Iethrisk7 = -0.3812495485312194500000000 /*7*/
+gen Iethrisk8 = -0.4064461381650994500000000 /*9*/ 
+gen Iethrisk9 = -0.2285715521377336100000000 /*6*/
+
+gen Ismoke0=0
+gen Ismoke1=0.2684479158158020200000000
+gen Ismoke2=0.6307674973877591700000000
+gen Ismoke3=0.7178078883378695700000000
+gen Ismoke4=0.8704172533465485100000000
+
+gen double dage = age
+replace dage=dage/10
+gen double age_1 = dage^(-1)
+gen double age_2 = dage^2
+gen double dbmi = bmi
+replace dbmi=dbmi/10
+gen double bmi_2 = ln(dbmi)*dbmi^(-2)
+gen double bmi_1 = dbmi^(-2)
+
+replace age_1 = age_1 - 0.233734160661697
+replace age_2 = age_2 - 18.304403305053711
+replace bmi_1 = bmi_1 - 0.146269768476486
+replace bmi_2 = bmi_2 - 0.140587374567986
+replace rati = rati - 4.321151256561279
+replace sbp = sbp - 130.589752197265620
+replace town = town - 0.551009356975555
+	
+
+
+gen double a=0
+
+/* The conditional sums */
+forvalues i=1/9 {
+replace a = a+ Iethrisk`i' if ethrisk==`i'
+} 
+forvalues i=0/4 {
+replace a = a+ Ismoke`i' if smoke_cat==`i' 
+}
+
+
+ replace a = a+ age_1 * -18.0437312550377270000000000 
+	 replace a = a+ age_2 * 0.0236486454254306940000000 
+	 replace a = a+ bmi_1 * 2.5388084343581578000000000 
+	 replace a = a+ bmi_2 * -9.1034725871528597000000000 
+	 replace a = a+ rati * 0.1684397636136909500000000 
+	 replace a = a+ sbp * 0.0105003089380754820000000 
+	 replace a = a+ town * 0.0323801637634487590000000 
+
+
+
+	/* Sum from boolean values */
+
+
+	 replace a = a+ b_AF * 1.0363048000259454000000000 
+	 replace a = a+ b_ra * 0.2519953134791012600000000 
+	 replace a = a+ b_renal * 0.8359352886995286000000000 
+	 replace a = a+ b_treatedhyp * 0.6603459695917862600000000 
+	 replace a = a+ b_type1 * 1.3309170433446138000000000 
+	 replace a = a+ b_type2 * 0.9454348892774417900000000 
+	 replace a = a+ fh_cvd * 0.5986037897136281500000000 
+
+
+
+	/* Sum from interaction terms */
+
+
+	 replace a = a+ age_1 * 1 * 0.6186864699379683900000000 if smoke_cat==1
+	 replace a = a+ age_1 * 1 * 1.5522017055600055000000000 if smoke_cat==2
+	 replace a = a+ age_1 * 1 * 2.4407210657517648000000000 if smoke_cat==3
+	 replace a = a+ age_1 * 1 * 3.5140494491884624000000000 if smoke_cat==4
+	 replace a = a+ age_1 * b_AF * 8.0382925558108482000000000 
+	 replace a = a+ age_1 * b_renal * -1.6389521229064483000000000 
+	 replace a = a+ age_1 * b_treatedhyp * 8.4621771382346651000000000 
+	 replace a = a+ age_1 * b_type1 * 5.4977016563835504000000000 
+	 replace a = a+ age_1 * b_type2 * 3.3974747488766690000000000 
+	 replace a = a+ age_1 * bmi_1 * 33.8489881012767600000000000 
+	 replace a = a+ age_1 * bmi_2 * -140.6707025404897100000000000 
+	 replace a = a+ age_1 * fh_cvd * 2.0858333154353321000000000 
+	 replace a = a+ age_1 * sbp * 0.0501283668830720540000000 
+	 replace a = a+ age_1 * town * -0.1988268217186850700000000 
+	 replace a = a+ age_2 * 1 * -0.0040893975066796338000000 if smoke_cat==1
+	 replace a = a+ age_2 * 1 * -0.0056065852346001768000000 if smoke_cat==2
+	 replace a = a+ age_2 * 1 * -0.0018261006189440492000000 if smoke_cat==3
+	 replace a = a+ age_2 * 1 * -0.0014997157296173290000000 if smoke_cat==4
+	 replace a = a+ age_2 * b_AF * 0.0052471594895864343000000 
+	 replace a = a+ age_2 * b_renal * -0.0179663586193546390000000 
+	 replace a = a+ age_2 * b_treatedhyp * 0.0092088445323379176000000 
+	 replace a = a+ age_2 * b_type1 * 0.0047493510223424558000000 
+	 replace a = a+ age_2 * b_type2 * -0.0048113775783491563000000 
+	 replace a = a+ age_2 * bmi_1 * 0.0627410757513945650000000 
+	 replace a = a+ age_2 * bmi_2 * -0.2382914909385732100000000 
+	 replace a = a+ age_2 * fh_cvd * -0.0049971149213281010000000 
+	 replace a = a+ age_2 * sbp * -0.0000523700987951435090000 
+	 replace a = a+ age_2 * town * -0.0012518116569283104000000 
+
+
+
+
+
+	/* Sum from continuous values */
+
+
+ gen double scorem = exp(a)
+ replace  scorem=survivor`surv'^scorem
+ replace  scorem = 1-scorem
+ replace scorem=100*scorem
+ *gen double scorem = 100.0 * (1 - survivor`surv'^(exp(a)))
+ 
+keep scorem patid rec_num
+
+tempfile temper
+save `temper' , replace
+restore
+merge 1:1 patid rec_num using `temper' , nogen
+drop rec_num
+
+replace score=scorem if scorem!=.
+drop scorem
+
+end
+ 
+
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
diff --git a/pr_cvd_male_raw_bmirep_qrisk3.do b/pr_cvd_male_raw_bmirep_qrisk3.do
new file mode 100644
index 0000000..dc623f5
--- /dev/null
+++ b/pr_cvd_male_raw_bmirep_qrisk3.do
@@ -0,0 +1,192 @@
+*use a dataset including 
+cap prog drop cvd_male_raw_qrisk3
+program define cvd_male_raw_qrisk3
+
+noi di "age -- age as integer"
+noi di	"b_AF -- atrial fibrilation as binary 01"
+noi di "b_ra -- rheumatoid arthritis as binary 01"
+noi di "b_renal -- rename disease as binary 01"
+noi di "b_treatedhyp -- hypertension treatment as binary 01"
+noi di "b_type1 -- type 1 diabetes as binary 01"
+noi di "b_type2 -- type 2 diabetes as binary 01"
+noi di "bmi -- bmi as continuous numeric, replaced with 20 if below 20"
+noi di "ethrisk -- ethnicity as categorical integer (0-9)"
+noi di "fh_cvd -- family history as binary 01"
+noi di "rati -- HDL LDL cholesterol ratio as continuous numeric"
+noi di "sbp -- systolic blood pressure as continuous numeric"
+noi di "smoke_cat -- smoking as integer category (0-4)"
+noi di "sex  -- sex as 1=male 2=female"
+
+noi di "b_atypicalantipsy -- atypical antipsychotic prescribing as binary 01"
+noi di "b_semi - severe mental illness as binary 01"
+noi di "sbp_s5 - systolic blood pressure variability as continuous numeric"
+noi di "b_impotence2 - erectile dysfunction as binary 01"
+noi di "b_sle - systemic lupus erythematosus as binary 01"
+noi di "b_corticosteroids - corticosteroid prescribing as binary 01"
+
+cap gen score=.
+gen double rec_num=_n
+preserve
+keep if sex==1
+*replace bmi=20 if bmi<20
+
+
+local surv=10
+forvalues i = 0/15 {
+	gen survivor`i'=0
+}
+replace survivor10=0.977268040180206
+*this part is added in incase they want to do risk scores over a different number of years but they haven't actually done the groundwork yet.
+
+gen Iethrisk0 = 0
+gen Iethrisk1 = 0 /*5*/
+gen Iethrisk2 = 0.2771924876030827900000000 /*3*/
+gen Iethrisk3 = 0.4744636071493126800000000 /*2*/
+gen Iethrisk4 = 0.5296172991968937100000000 /*1*/
+gen Iethrisk5 = 0.0351001591862990170000000 /*4*/
+gen Iethrisk6 = -0.3580789966932791900000000 /*8*/
+gen Iethrisk7 = -0.4005648523216514000000000 /*7*/
+gen Iethrisk8 = -0.4152279288983017300000000 /*9*/ 
+gen Iethrisk9 = -0.2632134813474996700000000 /*6*/
+
+gen Ismoke0=0
+gen Ismoke1=0.1912822286338898300000000
+gen Ismoke2=0.5524158819264555200000000
+gen Ismoke3=0.6383505302750607200000000
+gen Ismoke4=0.7898381988185801900000000
+
+gen double dage = age
+replace dage=dage/10
+gen double age_1 = dage^(-1)
+gen double age_2 = dage^3
+gen double dbmi = bmi
+replace dbmi=dbmi/10
+gen double bmi_2 = ln(dbmi)*dbmi^(-2)
+gen double bmi_1 = dbmi^(-2)
+
+replace age_1 = age_1 - 0.234766781330109
+replace age_2 = age_2 - 77.284080505371094
+replace bmi_1 = bmi_1 - 0.149176135659218
+replace bmi_2 = bmi_2 - 0.141913309693336
+replace rati = rati - 4.300998687744141
+replace sbp = sbp - 128.571578979492190
+replace sbps5 = sbps5 - 8.756621360778809
+replace town = town - 0.526304900646210
+	
+
+
+gen double a=0
+
+/* The conditional sums */
+forvalues i=1/9 {
+replace a = a+ Iethrisk`i' if ethrisk==`i'
+} 
+forvalues i=0/4 {
+replace a = a+ Ismoke`i' if smoke_cat==`i' 
+}
+
+
+ replace a = a+ age_1 * -17.8397816660055750000000000 
+	 replace a = a+ age_2 * 0.0022964880605765492000000 
+	 replace a = a+ bmi_1 * 2.4562776660536358000000000 
+	 replace a = a+ bmi_2 * -8.3011122314711354000000000 
+	 replace a = a+ rati * 0.1734019685632711100000000 
+	 replace a = a+ sbp * 0.0129101265425533050000000
+	 replace a = a+ sbps5 * 0.0102519142912904560000000 if sbps5 != .
+	 replace a = a+ town * 0.0332682012772872950000000 
+
+	/* Sum from boolean values */
+
+
+	 replace a = a+ b_AF * 0.8820923692805465700000000
+	 replace a = a+ b_atypicalantipsy * 0.1304687985517351300000000
+	 replace a = a+ b_corticosteroids * 0.4548539975044554300000000
+	 replace a = a+ b_impotence2 * 0.2225185908670538300000000
+	 replace a = a+ b_migraine * 0.2558417807415991300000000
+	 replace a = a+ b_ra * 0.2097065801395656700000000 
+	 replace a = a+ b_renal * 0.7185326128827438400000000
+	 replace a = a+ b_semi * 0.1213303988204716400000000
+	 replace a = a+ b_sle * 0.4401572174457522000000000
+	 replace a = a+ b_treatedhyp * 0.5165987108269547400000000 
+	 replace a = a+ b_type1 * 1.2343425521675175000000000 
+	 replace a = a+ b_type2 * 0.8594207143093222100000000 
+	 replace a = a+ fh_cvd * 0.5405546900939015600000000 
+
+
+
+	/* Sum from interaction terms */
+
+
+	 replace a = a+ age_1 * 1 * -0.2101113393351634600000000 if smoke_cat==1
+	 replace a = a+ age_1 * 1 * 0.7526867644750319100000000 if smoke_cat==2
+	 replace a = a+ age_1 * 1 * 0.9931588755640579100000000 if smoke_cat==3
+	 replace a = a+ age_1 * 1 * 2.1331163414389076000000000 if smoke_cat==4
+	 replace a = a+ age_1 * b_AF * 3.4896675530623207000000000 
+	 replace a = a+ age_1 * b_corticosteroids * 1.1708133653489108000000000
+	 replace a = a+ age_1 * b_impotence2 * -1.5064009857454310000000000
+	 replace a = a+ age_1 * b_migraine * 2.3491159871402441000000000
+	 replace a = a+ age_1 * b_renal * -0.5065671632722369400000000 
+	 replace a = a+ age_1 * b_treatedhyp * 6.5114581098532671000000000 
+	 replace a = a+ age_1 * b_type1 * 5.3379864878006531000000000 
+	 replace a = a+ age_1 * b_type2 * 3.6461817406221311000000000 
+	 replace a = a+ age_1 * bmi_1 * 31.0049529560338860000000000 
+	 replace a = a+ age_1 * bmi_2 * -111.2915718439164300000000000 
+	 replace a = a+ age_1 * fh_cvd * 2.7808628508531887000000000 
+	 replace a = a+ age_1 * sbp * 0.0188585244698658530000000 
+	 replace a = a+ age_1 * town * -0.1007554870063731000000000
+	 replace a = a+ age_2 * -0.0004985487027532612100000 if smoke_cat==1
+	 replace a = a+ age_2 * -0.0007987563331738541400000 if smoke_cat==2
+	 replace a = a+ age_2 * -0.0008370618426625129600000 if smoke_cat==3
+	 replace a = a+ age_2 * -0.0007840031915563728900000 if smoke_cat==4
+	 replace a = a+ age_2 * b_AF * -0.0003499560834063604900000
+	 replace a = a+ age_2 * b_corticosteroids * -0.0002496045095297166000000
+	 replace a = a+ age_2 * b_impotence2 * -0.0011058218441227373000000
+	 replace a = a+ age_2 * b_migraine * 0.0001989644604147863100000
+	 replace a = a+ age_2 * b_renal * -0.0018325930166498813000000 
+	 replace a = a+ age_2 * b_treatedhyp * 0.0006383805310416501300000 
+	 replace a = a+ age_2 * b_type1 * 0.0006409780808752897000000 
+	 replace a = a+ age_2 * b_type2 * -0.0002469569558886831500000 
+	 replace a = a+ age_2 * bmi_1 * 0.0050380102356322029000000 
+	 replace a = a+ age_2 * bmi_2 * -0.0130744830025243190000000 
+	 replace a = a+ age_2 * fh_cvd * -0.0002479180990739603700000 
+	 replace a = a+ age_2 * sbp * -0.0000127187419158845700000 
+	 replace a = a+ age_2 * town * -0.0000932996423232728880000 
+
+
+
+
+
+	/* Sum from continuous values */
+
+
+ gen double scorem = exp(a)
+ replace  scorem=survivor`surv'^scorem
+ replace  scorem = 1-scorem
+ replace scorem=100*scorem
+ *gen double scorem = 100.0 * (1 - survivor`surv'^(exp(a)))
+ 
+keep scorem patid rec_num
+
+tempfile temper
+save `temper' , replace
+restore
+merge 1:1 patid rec_num using `temper' , nogen
+drop rec_num
+
+replace score=scorem if scorem!=.
+drop scorem
+
+end
+ 
+
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
diff --git a/pr_get_ED.do b/pr_get_ED.do
new file mode 100644
index 0000000..2fed121
--- /dev/null
+++ b/pr_get_ED.do
@@ -0,0 +1,29 @@
+cap prog drop pr_get_ED
+
+prog define pr_get_ED 
+syntax , clinicalfile(string) referralfile(string) therapyfile(string) begin(string) end(string) runin(string) [wide]
+
+preserve
+*first ever diagnosis code
+pr_get_bclin, variable(ED) qof(notqof) clinicalfile(`clinicalfile') referralfile(`referralfile') begin(`begin') end(`end') runin(`runin')
+
+tempfile b_ED
+save `b_ED', replace
+
+restore
+*first ever treatment
+pr_get_btherapy, variable(ED_drugs) therapyfile(`therapyfile') begin(`begin') end(`end') runin(`runin') time(ever)
+
+append using `b_ED'
+
+*first of either diagnosis or treatment
+replace EDdate = ED_drugsdate if EDdate == .
+assert EDdate !=.
+
+replace b_ED = b_ED_drugs if b_ED == .
+assert b_ED !=.
+
+drop ED_drugs b_ED_drugs
+bysort patid EDdate: keep if _n==1
+
+end
diff --git a/pr_get_bclin.do b/pr_get_bclin.do
new file mode 100644
index 0000000..1175191
--- /dev/null
+++ b/pr_get_bclin.do
@@ -0,0 +1,74 @@
+cap prog drop pr_get_bclin
+
+capture program drop pr_get_bclin
+prog define pr_get_bclin
+syntax , variable(string) qof(string) clinicalfile(string) referralfile(string) begin(string) end(string) runin(string) [wide]
+
+
+preserve
+
+*get all clinical and referral events matching code list
+if "`qof'" == "qof" {
+	use "$QRiskCodelistdir\cr_codelist_qof_cod.dta", clear
+	keep if variable=="b_`variable'"
+	}
+	else {
+		use "$QRiskCodelistdir\cr_codelist_`variable'.dta", clear
+		}
+
+merge 1:m medcode using `clinicalfile' , nogen keep(match)
+
+tempfile temp
+save `temp', replace
+
+if "`qof'" == "qof" {
+	use "$QRiskCodelistdir\cr_codelist_qof_cod.dta", clear
+	keep if variable=="b_`variable'"
+	}
+	else {
+		use "$QRiskCodelistdir\cr_codelist_`variable'.dta", clear
+		}
+
+merge 1:m medcode using `referralfile' , nogen keep(match)
+append using `temp'
+
+gen b_`variable'=1
+keep patid eventdate b_`variable'
+tempfile bclin
+save `bclin' , replace
+
+*merge with patient file
+restore
+merge 1:m patid using `bclin', keep(match master)
+
+*keep first event within specified time table
+drop if _merge==3 & eventdate>`end'
+drop if _merge==3 & eventdate<`begin'-365.25*`runin'
+gsort patid -eventdate
+duplicates drop patid if eventdate<`begin' & _merge==3 , force
+
+sort patid eventdate
+by patid: keep if _n==1
+
+replace b_`variable'=0 if b_`variable'==.
+
+*add record for unexposed time between start of follow-up and first event
+expand 2 if eventdate>`begin' & eventdate!=. & patid[_n]!=patid[_n-1]
+sort patid eventdate
+by patid: replace b_`variable'=0 if _n==1 & eventdate>`begin'
+by patid: replace eventdate = `begin' if _n==1 & eventdate>`begin'
+
+drop _merge
+rename eventdate `variable'date
+
+*HS - I'm not sure what this loop is meant to do but it doesn't run in this do file
+if "`wide'"=="wide" {
+by patid: gen b_clin_num=_n
+
+reshape wide b_`variable' `variable'date, i(patid) j(b_clin_num)
+}
+
+end
+
+
+
diff --git a/pr_get_fh_cvd_60.do b/pr_get_fh_cvd_60.do
new file mode 100644
index 0000000..7e9dc61
--- /dev/null
+++ b/pr_get_fh_cvd_60.do
@@ -0,0 +1,37 @@
+cap prog drop pr_get_fh_cvd
+program define pr_get_fh_cvd
+syntax, clinicalfile(string) begin(string) end(string) runin(integer)
+
+preserve
+
+use `clinicalfile' , clear
+merge m:1 medcode using  "$QRiskCodelistdir\cr_codelist_fh_cvd_60" , nogen keep(match)
+rename eventdate fh_date
+keep fh_date patid
+
+tempfile tempura
+save `tempura' , replace
+
+restore
+merge 1:m patid using `tempura', nogen keep(match master)
+
+drop if fh_date<`begin'-`runin'*365.25 | fh_date>`end' & fh_date!=.
+sort patid fh_date
+by patid: keep if _n==1
+*keep first in study period*
+gen fh_cvd=1 if fh_date<=`begin'
+*set record to 1 if fh recorded before begin of follow up
+
+expand 2 if fh_cvd!=1 & fh_date!=.
+*generate another record if it has been recorded later in fup
+sort patid
+by patid: replace fh_date=`begin' if _N>1 & _n==1
+*first record keep fh=0 & set date to start of follow up
+by patid: replace fh_cvd=1 if _N>1 & _n==2
+*second record keep record date and set fh=1
+
+*set remaining missing records to zero
+replace fh_cvd=0 if fh_cvd==.
+replace fh_date=`begin' if fh_date==.
+
+end
diff --git a/pr_get_tchdl_medcoded.do b/pr_get_tchdl_medcoded.do
new file mode 100644
index 0000000..71540b1
--- /dev/null
+++ b/pr_get_tchdl_medcoded.do
@@ -0,0 +1,52 @@
+cap prog drop pr_get_tchdl_updates
+program define pr_get_tchdl_updates
+syntax, testfile(string) begin(string) end(string) runin(integer) [wide cutoff(string)]
+
+if "`cutoff'"=="" {
+	local cutoff "."
+}
+
+preserve
+
+use `testfile' , clear
+
+keep if medcode==14371 | medcode==14372 | medcode==14108
+keep if data3==0 | data3==1 | data3==151 | data3==161
+drop if data2>`cutoff'
+drop if data2==0
+rename data2 TC_HDLratio
+rename eventdate TC_HDLdate
+drop if TC_HDLdate == .
+
+*HDL:LDL ratios recorded on same date
+keep patid TC_HDLratio TC_HDLdate
+duplicates drop
+duplicates tag patid TC_HDLdate, gen(dup)
+drop if dup > 0 /*Different lab tests on same day indicative of error?*/
+
+keep patid TC_HDLratio TC_HDLdate
+tempfile tempura
+save `tempura' , replace
+restore
+merge 1:m patid using `tempura' , nogen keep(match master)
+
+drop if (TC_HDLdate<`begin'-365.25*`runin' | TC_HDLdate>`end') & TC_HDLdate!=.
+gsort patid -TC_HDLdate
+duplicates drop patid if TC_HDLdate<`begin', force
+
+sort patid TC_HDLdate TC_HDLratio
+expand 2 if TC_HDLdate>`begin' & TC_HDLdate!=. & patid[_n]!=patid[_n-1]
+sort patid TC_HDLdate
+by patid: replace TC_HDLratio=. if _n==1 & TC_HDLdate>`begin'
+by patid: replace TC_HDLdate=`begin' if _n==1 & TC_HDLdate>`begin'
+
+
+if "`wide'"=="wide" {
+by patid: gen TC_HDLnum=_n
+reshape wide TC_HDLratio TC_HDLdate, i(patid) j(TC_HDLnum)
+}
+
+
+end
+
+
diff --git a/pr_getdiabetes.do b/pr_getdiabetes.do
new file mode 100644
index 0000000..9ea1630
--- /dev/null
+++ b/pr_getdiabetes.do
@@ -0,0 +1,80 @@
+cap prog drop pr_get_diabetes
+
+capture program drop pr_get_diabetes
+prog define pr_get_diabetes
+syntax , clinicalfile(string) referralfile(string) begin(string) end(string) runin(string) [wide]
+
+
+preserve
+
+*get all clinical and referral events matching code list
+use "$QRiskCodelistdir\cr_codelist_qof_cod.dta", clear
+keep if variable == "b_type1" | variable == "b_type2"
+
+merge 1:m medcode using `clinicalfile' , nogen keep(match)
+
+tempfile temp
+save `temp', replace
+
+use "$QRiskCodelistdir\cr_codelist_qof_cod.dta", clear
+keep if variable == "b_type1" | variable == "b_type2"
+
+merge 1:m medcode using `referralfile' , nogen keep(match)
+append using `temp'
+
+gen diabetes = 1 if variable == "b_type1"
+replace diabetes = 2 if variable == "b_type2"
+keep patid eventdate diabetes
+
+tempfile bdiab
+save `bdiab' , replace
+
+*merge with patient file
+restore
+merge 1:m patid using `bdiab', keep(match master)
+
+*keep records within specified timetable
+drop if _merge==3 & eventdate>`end'
+drop if _merge==3 & eventdate<`begin'-365.25*`runin'
+gsort patid -eventdate
+duplicates drop patid if eventdate<`begin' & _merge==3 , force
+
+sort patid eventdate
+*by patid: keep if _n==1
+
+replace diabetes=0 if diab==.
+
+*if sequential records are for the same diabetes type, keep the first record
+by patid: gen drop = 1 if patid[_n]==patid[_n-1] & diab[_n] == diab[_n-1]
+drop if drop == 1
+drop drop
+
+*add record for unexposed time between start of follow-up and first event
+expand 2 if eventdate>`begin' & eventdate!=. & patid[_n]!=patid[_n-1]
+sort patid eventdate
+by patid: replace diab=0 if _n==1 & eventdate>`begin'
+by patid: replace eventdate = `begin' if _n==1 & eventdate>`begin'
+
+/*at each record, set the recorded type to 1 and the other to 0
+i.e. diabetes is defined by the latest record and patients cannot have
+type 1 and type 2 diabetes at the same time*/
+gen b_type1 = 0
+gen b_type2 = 0
+replace b_type1 = 1 if diab == 1
+replace b_type2 = 1 if diab == 2
+
+gen type1date = eventdate
+gen type2date = eventdate
+format type1date type2date %dD/N/CY
+drop eventdate diabetes
+
+if "`wide'"=="wide" {
+by patid: gen b_clin_num=_n
+
+reshape wide b_`variable' `variable'date, i(patid) j(b_clin_num)
+}
+
+end
+
+
+
diff --git a/pr_getlastsbpvariability.do b/pr_getlastsbpvariability.do
new file mode 100644
index 0000000..b6303a5
--- /dev/null
+++ b/pr_getlastsbpvariability.do
@@ -0,0 +1,60 @@
+cap prog drop pr_getlastsbpvariability
+
+program define pr_getlastsbpvariability 
+syntax , clinicalfile(string) additionalfile(string)   index(string) runin(integer) [limitlow(string) limithigh(string)]
+
+if "`limitlow'"=="" {
+	local limitlow "0"
+}
+
+if "`limithigh'"=="" {
+	local limithigh "."
+}
+preserve
+
+use "$QRiskCodelistdir\cr_codelist_qof_cod.dta", clear
+keep if variable=="sbp"
+merge 1:m medcode using  `clinicalfile' , keep(match) nogen
+
+replace adid = -_n if adid==0
+rename value sbp
+merge 1:1 patid adid using `additionalfile',  keep(match master)
+
+replace sbp=data2 if data2!=.
+
+drop if eventdate == .
+drop if sbp==.
+drop if sbp==1
+drop if sbp >`limithigh'
+drop if sbp <`limitlow'
+
+*keep lowest sbp if more than one measurement on same day
+keep patid eventdate sbp
+duplicates tag patid eventdate, gen(dup)
+bysort patid eventdate (sbp): keep if _n==1
+
+keep patid eventdate sbp
+
+tempfile sbprecs
+save `sbprecs' , replace
+
+
+restore
+merge 1:m patid using `sbprecs' ,  keep(match master)
+sort patid eventdate
+
+
+*identify all sbp measurements recorded in the five years before study entry
+gen _diff = `index' - eventdate
+gen _sbp5yr = 1 if _diff < (365.25*5) & _diff >= 0
+by patid: egen _countsbp5yr = count(_sbp5yr)
+
+*calculate standard deviation where there are two or more recorded values
+by patid: egen sbp_sd = sd(sbp) if _sbp5yr == 1 & _countsbp5yr >= 2
+keep patid sbp_sd
+keep if sbp_sd !=.
+duplicates drop
+isid patid
+
+end
+
diff --git a/pr_getsbpupdates.do b/pr_getsbpupdates.do
new file mode 100644
index 0000000..0c6a64c
--- /dev/null
+++ b/pr_getsbpupdates.do
@@ -0,0 +1,67 @@
+cap prog drop pr_getsbpupdates
+
+program define pr_getsbpupdates
+syntax , clinicalfile(string) additionalfile(string)  begin(string) end(string) runin(integer) [limitlow(string) limithigh(string)] [wide]
+
+if "`limitlow'"=="" {
+	local limitlow "0"
+}
+
+if "`limithigh'"=="" {
+	local limithigh "."
+}
+
+preserve
+
+use "$QRiskCodelistdir\cr_codelist_qof_cod", clear
+keep if variable=="sbp"
+merge 1:m medcode using  `clinicalfile' , keep(match) nogen
+
+replace adid = -_n if adid==0
+rename value sbp
+merge 1:1 patid adid using `additionalfile',  keep(match master)
+replace sbp=data2 if data2!=.
+
+drop if eventdate == .
+drop if sbp==.
+drop if sbp==1
+drop if sbp >`limithigh'
+drop if sbp <`limitlow'
+
+keep patid eventdate sbp
+
+*keep lowest sbp if more than one measurement on same day
+keep patid eventdate sbp
+duplicates tag patid eventdate, gen(dup)
+bysort patid eventdate (sbp): keep if _n==1
+
+tempfile sbprecs
+save `sbprecs' , replace
+
+restore
+merge 1:m patid using `sbprecs' ,  keep(match master)
+
+rename eventdate bpdate
+drop if (bpdate<`begin'-365.25*`runin' | bpdate>`end') & _merge==3
+gsort patid -bpdate
+duplicates drop patid if bpdate<`begin', force
+
+sort patid bpdate
+by patid: drop if sbp[_n-1]==sbp & _n>1
+
+expand 2 if bpdate>`begin' & bpdate!=. & patid[_n]!=patid[_n-1]
+sort patid bpdate
+by patid: replace sbp=. if _n==1 & bpdate>`begin'
+by patid: replace bpdate=`begin' if _n==1 & bpdate>`begin'
+
+
+
+drop _merge
+if "`wide'"=="wide" {
+by patid: gen sbp_rec_num=_n
+
+reshape wide sbp bpdate, i(patid) j(sbp_rec_num)
+}
+
+end
+
diff --git a/pr_getsmokingupdates.do b/pr_getsmokingupdates.do
new file mode 100644
index 0000000..aaa9cb5
--- /dev/null
+++ b/pr_getsmokingupdates.do
@@ -0,0 +1,103 @@
+cap prog drop pr_getsmokingupdates
+
+program define pr_getsmokingupdates 
+syntax , clinicalfile(string) additionalfile(string)  smokingstatusvar(string) begin(string) end(string) runin(integer) [wide]
+
+preserve
+
+use "$QRiskCodelistdir\cr_codelist_qof_cod", clear
+keep if variable== "smoke_cat"
+*NOTE: EH AMENDED 21/09/2018 - MEDCODE 54, Read code 137..00 "Tobacco consumption" is being removed from the code list as it does not indicate smoking status.
+drop if medcode==54
+merge 1:m medcode using  `clinicalfile' , keep(match) nogen
+
+rename value clinstatus
+label define clinstatuslab 0 "non-smoker" 1 "ex-smoker" 2 "light smoker" 3 "moderate smoker" 4 "heavy smoker" 234 "smoker amount unknown", replace
+label values clinstatus clinstatuslab
+
+*merge with additional file
+*main medcodes associated with enttype that are not in the smoking code list = 54 tobacco consumption, 60 current non-smoker, 2111 health education - smoking, 7622 smoking cessation advice
+replace adid = -_n if adid==0
+merge 1:1 patid adid using `additionalfile'
+drop if eventdate == .
+
+drop if _merge == 2 & enttype !=4 /*drop additional files not found using codelist and not in the smoking entity type*/
+replace clinstatus = . if _merge == 2
+tab medcode _merge, m
+
+*smoking status using additional clinical details file
+gen adstatus = .
+replace adstatus = 0 if data1 == 2
+replace adstatus = 1 if data1 == 3
+replace adstatus = 234 if data1 == 1
+label values adstatus clinstatuslab
+
+*update clinical and additional details status using cigarettes per day
+rename data2 cigsperday
+replace cigsperday = 0 if cigsperday==0
+replace cigsperday = 2 if cigsperday<10
+replace cigsperday = 3 if cigsperday<20 & cigsperday>9
+replace cigsperday = 4 if cigsperday>19 & cigsperday!=.
+label values cigsperday clinstatuslab
+
+foreach file in ad clin {
+	replace `file'status=2 if inlist(`file'status,234,2,3,4)  & cigsperday==2
+	replace `file'status=3 if inlist(`file'status,234,2,3,4) & cigsperday==3
+	replace `file'status=4 if inlist(`file'status,234,2,3,4) & cigsperday==4
+	gen `file's234 = 1 if `file'status==234
+	replace `file'status=3 if `file'status==234 /*assume moderate smoker if no info*/
+	*replace status=2 if status==1234
+	}
+
+*prioritise additional clinical details status unless missing or information about amount smoked more complete in additional clinical details file
+tab clinstatus adstatus, m
+gen status = clinstatus
+replace status = adstatus if clinstatus == .
+label values status clinstatuslab 
+gen s234 = 1 if clins234 == 1
+replace s234 = 1 if clinstatus == . 
+
+*duplicates
+duplicates drop patid eventdate status, force
+duplicates tag patid eventdate, gen(dup)
+sort patid eventdate
+by patid eventdate: egen counts234 = count(s234)
+by patid eventdate: egen minstatus = min(status)
+by patid eventdate: egen maxstatus = max(status)
+drop if s234 == 1 & counts234 <= dup & dup> 0 & maxstatus > 2 & minstatus > 1 /*keep specified amount if imputed amount on same day*/
+drop dup
+duplicates tag patid eventdate, gen(dup)
+drop if dup > 0 /*drop all remaining records with at least two different smoking status records on the same day*/
+assert status !=.
+
+keep patid eventdate status
+
+tempfile smokrecs
+save `smokrecs' , replace
+
+restore
+merge 1:m patid using `smokrecs' ,  keep(match master)
+
+rename eventdate smoke_update
+drop if (smoke_update<`begin'-365.25*`runin' | smoke_update>`end') & _merge==3
+gsort patid -smoke_update
+duplicates drop patid if smoke_update<`begin', force
+
+sort patid smoke_update
+by patid: drop if status[_n-1]==status & _n>1
+
+expand 2 if smoke_update>`begin' & smoke_update!=. & patid[_n]!=patid[_n-1]
+sort patid smoke_update
+by patid: replace status=. if _n==1 & smoke_update>`begin'
+by patid: replace smoke_update=`begin' if _n==1 & smoke_update>`begin'
+
+rename status `smokingstatusvar'
+drop _merge
+if "`wide'"=="wide" {
+by patid: gen smok_rec_num=_n
+
+reshape wide `smokingstatusvar' smoke_update, i(patid) j(smok_rec_num)
+}
+
+end
+
diff --git a/pr_getsmokingupdates_qrisk2.do b/pr_getsmokingupdates_qrisk2.do
new file mode 100644
index 0000000..17bae82
--- /dev/null
+++ b/pr_getsmokingupdates_qrisk2.do
@@ -0,0 +1,55 @@
+cap prog drop pr_getsmokingupdates
+
+program define pr_getsmokingupdates 
+syntax , clinicalfile(string) additionalfile(string)  smokingstatusvar(string) begin(string) end(string) runin(integer) [wide]
+
+preserve
+
+use "$QRiskCodelistdir\cr_codelist_qof_cod", clear
+keep if variable== "smoke_cat"
+*NOTE: EH AMENDED 21/09/2018 - MEDCODE 54, Read code 137..00 "Tobacco consumption" is being removed from the code list as it does not indicate smoking status.
+drop if medcode==54
+merge 1:m medcode using  `clinicalfile' , keep(match) nogen
+
+replace adid = -_n if adid==0
+rename value status
+merge 1:1 patid adid using `additionalfile',  keep(match master)
+rename data2 cigsperday
+replace status=2 if status==2 & cigsperday<10
+replace status=3 if status==2 & cigsperday<20 & cigsperday>9
+replace status=4 if status==2 & cigsperday>19 & cigsperday!=.
+
+replace status=3 if status==234
+replace status=2 if status==1234
+
+keep patid eventdate status
+
+tempfile smokrecs
+save `smokrecs' , replace
+
+restore
+merge 1:m patid using `smokrecs' ,  keep(match master)
+
+rename eventdate smoke_update
+drop if (smoke_update<`begin'-365.25*`runin' | smoke_update>`end') & _merge==3
+gsort patid -smoke_update
+duplicates drop patid if smoke_update<`begin', force
+
+sort patid smoke_update
+by patid: drop if status[_n-1]==status & _n>1
+
+expand 2 if smoke_update>`begin' & smoke_update!=. & patid[_n]!=patid[_n-1]
+sort patid smoke_update
+by patid: replace status=. if _n==1 & smoke_update>`begin'
+by patid: replace smoke_update=`begin' if _n==1 & smoke_update>`begin'
+
+rename status `smokingstatusvar'
+drop _merge
+if "`wide'"=="wide" {
+by patid: gen smok_rec_num=_n
+
+reshape wide `smokingstatusvar' smoke_update, i(patid) j(smok_rec_num)
+}
+
+end
+